Hybrid system networked converter long time domain predictive control and non-disturbance switching method

Abstract

The application discloses a hybrid system network converter long-time domain prediction control and non-disturbance switching method, and belongs to the technical field of power system control. In view of the problems of poor dynamic performance of a hybrid system of a network construction / following network converter under high proportion of new energy grid connection and easy impact of mode switching, the method is as follows: a long-time domain model is constructed to predict current control, a traditional PI current inner ring is replaced, a dq coordinate system is discretized to model, double-step length prediction is compensated to delay, a Lagrange extrapolation is used to update a reference current, current tracking accuracy and harmonic suppression capability are improved; a non-disturbance switching strategy is designed, an alpha-beta coordinate system and a current control ring are unified, a switching angle is locked and an initial value of an integrator is synchronized, a reference voltage is dynamically adjusted to guarantee variable continuity; the two strategies are integrated into a hybrid framework, stable operation and smooth switching under strong / weak power grid and transition working conditions are realized. The application reduces grid-connected current total harmonic distortion rate to below 0.85%, has no voltage and current impact in switching, and has the advantages of enhancing power system stability and flexibility.

Description

Technical Field

[0001] This invention belongs to the field of power system control technology, specifically relating to a long-time-domain predictive control and disturbance-free switching method for a hybrid system grid converter. Background Technology

[0002] With the increasing penetration rate of new energy sources, the inertia and voltage support capabilities of power systems have significantly decreased. The safe and stable operation of hybrid systems combining grid-forming converters (GFM) and grid-following converters (GFL) has gradually become a research hotspot. GFM can actively establish voltage and frequency references, providing system support; GFL relies on phase-locked loops (PLLs) to track grid signals and achieve power point tracking. In scenarios of large-scale grid connection of new energy sources, how to achieve coordinated operation of grid-forming and grid-following equipment, and improve the dynamic control performance of grid-forming equipment, has become a critical issue that urgently needs to be addressed. Virtual synchronous generator control, as a typical grid-forming strategy, can simulate the inertia and damping characteristics of synchronous machines. However, it typically uses a PI current inner loop, which has problems such as limited dynamic response speed and insufficient harmonic suppression capability, resulting in system frequency being easily affected by disturbances and experiencing large fluctuations, as well as a high harmonic distortion rate of the grid-connected current. Furthermore, the switching between grid-forming and grid-following modes in hybrid systems can easily trigger voltage and current surges, affecting the stable operation of the system. Therefore, a control method is needed that can improve current tracking accuracy, enhance frequency stability, reduce harmonic distortion rate, and achieve smooth, disturbance-free switching between grid-connected and grid-following modes in order to address the challenges brought about by a high proportion of new energy grid connection. Summary of the Invention

[0003] To address the above problems, this invention provides a long-time-domain predictive control and disturbance-free switching method for grid-connected converters in hybrid systems. This method improves dynamic response and steady-state performance by introducing long-time-domain model predictive control to replace the traditional PI current inner loop, and designs a disturbance-free switching strategy to achieve a smooth transition between grid-connected and grid-following modes, effectively enhancing system stability, power quality, and operational flexibility.

[0004] The technical solution adopted in this invention is as follows: A method for long-time domain predictive control and disturbance-free switching of a grid-connected converter in a hybrid system, characterized by comprising the following steps:

[0005] S1. Construct a long-time-domain model predictive current control strategy for grid-type converters to replace the current PI inner loop in traditional virtual synchronous generator control:

[0006] S11. Based on the dynamic equation of the output current in the synchronously rotating dq coordinate system, the discrete state-space model is obtained by discretization using the forward Euler method.

[0007] S12. Design a long-time domain prediction model, and compensate for system delay by predicting the current value in the next two time steps and applying different voltage vectors.

[0008] S13. The reference current is updated in real time using the second-order Lagrange extrapolation method to provide an accurate tracking target for predictive control.

[0009] S14. Solve the value function through rolling optimization, enumerate all possible voltage vectors and select the voltage vector that minimizes the value function as the optimal control quantity;

[0010] S2. Design a disturbance-free switching control strategy between grid-type and follow-grid-type converters:

[0011] S21. Establish a control principle model for a grid-type converter, transform the terminal voltage coordinate system in the grid-type converter mode to the αβ coordinate system, realize the same coordinate system as the grid-type converter mode, and share the current control loop between the two modes.

[0012] S22. In grid-type converter mode, the required angle for grid-type converter mode is calculated in real time using the power synchronization loop. The angle is locked during switching because the switching occurs in grid-connected mode, and the active power command controlled by the grid-connected converter and the grid-connected converter is... When set to the same value, the angular velocity output by the power synchronization loop and the angular velocity output by the grid-type control unit remain consistent under steady state.

[0013] S23. When switching from a grid-type converter to a follow-grid type converter, the grid synchronization part changes from a power synchronization loop to a phase-locked loop. Set the initial value for the phase-locked loop integrator; when switching from a grid-type to a grid-type converter, the grid synchronization part changes from a phase-locked loop to a power synchronization loop. Set the initial value for the power synchronization loop integrator;

[0014] S24. During the switching process, the reference voltage value is dynamically adjusted based on the difference between the internal potential and the terminal voltage. The reference voltage value is modulated by SPWM to obtain the PWM duty cycle, and then the PWM control signal is dynamically adjusted to keep the variables continuous.

[0015] S3. Integrate long-time-domain model predictive control strategy and disturbance-free switching strategy into the hybrid system control framework. Implement long-time-domain model predictive current control in grid-type converters and configure disturbance-free switching control mechanism to achieve stable operation and smooth mode switching of the system under strong grid, weak grid and transition conditions.

[0016] Furthermore, in step S1, the predictive control model of the grid-type converter is shown in formulas (1)-(6), the forward Euler method is shown in formula (7), and the model is discretized. The discrete state space model is shown in formulas (8)-(9), the single-step prediction model is shown in formulas (10)-(11), the long-time domain prediction model is shown in formulas (12)-(13), the value function is shown in formula (14), the second-order Lagrange extrapolation algorithm is shown in formula (15), and the derivation of the reference current values ​​for the next two sampling times is shown in formulas (16)-(17).

[0017] (1)

[0018] (2)

[0019] (3)

[0020] (4)

[0021] (5)

[0022] (6)

[0023] (7)

[0024] (8)

[0025] (9)

[0026] (10)

[0027] (11)

[0028] (12)

[0029] (13)

[0030] (14)

[0031] (15)

[0032] (16)

[0033] (17)

[0034] In the formula: These are the switching signals for a three-phase inverter, representing each switching state. , , It can be 0 or 1; It is a DC voltage; Let be the output voltage vector, where , , The value is Or it could be 0; For grid-connected current Axial components; For grid-connected current Axial components; For converter Shaft output voltage; For converter Shaft output voltage; For grid voltage Axial components; For grid voltage Axial components; , These are the filter inductor and resistor, respectively. The system angular velocity; This is the sampling period of the system, i.e., the time of a single step. For the first Each sampling period Predicted shaft current value; For the first Each sampling period Predicted shaft current value; For the first Each sampling period Shaft current reference value; For the first Each sampling period Shaft current reference value; For the first Each sampling period Predicted shaft current value; For the first Each sampling period Predicted shaft current value; For the first Each sampling period Shaft current reference value; For the first Each sampling period Shaft current reference value.

[0035] Furthermore, in step S2, a control principle model for the grid-type converter is established, as shown in formulas (18)-(22). To achieve accurate tracking of reactive power, an integral element is introduced into the control loop, as shown in formula (23). The coordinate system transformation is shown in formula (24). The reference current of the grid-type converter is shown in formulas (25)-(26). The reference current of the grid-type converter is shown in formulas (27)-(28). The reference voltage signal is calculated as shown in formulas (29)-(31).

[0036] (18)

[0037] (19)

[0038] (20)

[0039] (twenty one)

[0040] (twenty two)

[0041] (twenty three)

[0042] (twenty four)

[0043] (25)

[0044] (26)

[0045] (27)

[0046] (28)

[0047] (29)

[0048] (30)

[0049] (31)

[0050] In the formula: It is the equivalent moment of inertia; The system angular velocity; Rated angular velocity; This is the equivalent input power; Electromagnetic power; The damping coefficient; Rated active power; This is the active power droop coefficient; Rated voltage; This is the reactive power droop coefficient; Rated reactive power; For rotation angle; in GFM mode, it is... In GFL mode, ; , These are the proportional-integral adjustment parameters; For network mode Shaft reference current value; For network mode Shaft reference current value; For network construction mode Shaft reference current value; The internal potential of the system; Terminal voltage; Terminal current; For network construction mode Shaft reference current value; To generate the reference PWM signal Shaft voltage value; To generate the reference PWM signal Shaft voltage value; , Each represents the current time. Actual output value of shaft current; , , The three-phase voltage values ​​are used to generate the reference PWM signal.

[0051] The present invention also provides a long-time-domain predictive control and disturbance-free switching system for a hybrid system grid converter, comprising: a long-time-domain model predictive control module, a disturbance-free switching control module, and a hybrid system integrated control module;

[0052] The long-time-domain model predictive control module is configured to construct a long-time-domain model predictive current control strategy for a grid-type converter, replacing the current PI inner loop in the traditional virtual synchronous generator control. The long-time-domain model predictive control module includes a discretization modeling unit, a long-time-domain prediction unit, a reference current update unit, and an optimization solution unit. The discretization modeling unit, based on the output current dynamic equation in the synchronous rotating dq coordinate system, uses the forward Euler method for discretization to establish a discrete state-space model of the grid-type converter. The long-time-domain prediction unit designs a long-time-domain prediction model, predicting the current values ​​for the next two time steps and applying different voltage vectors to compensate for system delays. The reference current update unit uses a second-order Lagrange extrapolation algorithm to update the reference current in real time, providing an accurate tracking target for predictive control. The optimization solution unit solves the value function through rolling optimization, enumerating all possible voltage vectors and selecting the voltage vector that minimizes the value function as the optimal control quantity.

[0053] The disturbance-free switching control module is configured to achieve disturbance-free switching between grid-type and follow-grid converters. It includes a coordinate and control loop unification unit, a switching angle calculation and locking unit, a synchronous switching unit, and a variable coordination control unit. The coordinate and control loop unification unit establishes a control principle model for the grid-type converter, transforming the terminal voltage coordinate system in the follow-grid converter mode to the αβ coordinate system, achieving coordinate system unification with the grid-type converter mode, and enabling both modes to share the current control loop. The switching angle calculation and locking unit, in the grid-type converter mode, uses the power synchronization loop to calculate the required angle for the follow-grid converter mode in real time. And lock the angle during switching; the synchronous switching unit, when switching from a grid configuration to a follow grid configuration, controls the grid synchronization part to switch from a power synchronization loop to a phase-locked loop, and... Set the initial value for the phase-locked loop integrator; when switching from a follow-the-grid type to a multi-grid type, the control grid synchronization part switches from a phase-locked loop to a power synchronization loop, and... The initial value is set as the power synchronization loop integrator; the variable coordination control unit coordinates and controls the internal potential, terminal voltage, and terminal current during the switching process, calculates the reference PWM control signal, and ensures that the key variables are continuous without sudden changes.

[0054] The hybrid system integrated control module is configured to integrate long-time domain model predictive control strategy and disturbance-free switching strategy into the hybrid system control framework, so as to realize stable operation and smooth mode switching of the system under strong grid and weak grid conditions. The long-time domain model predictive control focuses on improving the dynamic performance and steady-state accuracy of grid-type converters, while the disturbance-free switching control ensures the smoothness of system operation mode transition.

[0055] Furthermore, in the discretization modeling unit, the predictive control model of the grid-type converter satisfies formulas (1)-(6), the forward Euler method satisfies formula (7), and the discrete state-space model satisfies formulas (8)-(9); in the long-time domain prediction unit, the single-step prediction model satisfies formulas (10)-(11), and the long-time domain prediction model satisfies formulas (12)-(13); in the optimization solution unit, the value function satisfies formula (14); in the reference current update unit, the second-order Lagrange extrapolation algorithm satisfies formula (15), and the derivation of the reference current value for the next two sampling times satisfies formulas (16)-(17).

[0056] (1)

[0057] (2)

[0058] (3)

[0059] (4)

[0060] (5)

[0061] (6)

[0062] (7)

[0063] (8)

[0064] (9)

[0065] (10)

[0066] (11)

[0067] (12)

[0068] (13)

[0069] (14)

[0070] (15)

[0071] (16)

[0072] (17)

[0073] In the formula: These are the switching signals for a three-phase inverter, representing each switching state. , , It can be 0 or 1; It is a DC voltage; Let be the output voltage vector, where , , The value is Or it could be 0; For grid-connected current Axial components; For grid-connected current Axial components; For converter Shaft output voltage; For converter Shaft output voltage; For grid voltage Axial components; For grid voltage Axial components; , These are the filter inductor and resistor, respectively. The system angular velocity; This is the sampling period of the system, i.e., the time of a single step. For the first Each sampling period Predicted shaft current value; For the first Each sampling period Predicted shaft current value; For the first Each sampling period Shaft current reference value; For the first Each sampling period Shaft current reference value; For the first Each sampling period Predicted shaft current value; For the first Each sampling period Predicted shaft current value; For the first Each sampling period Shaft current reference value; For the first Each sampling period Shaft current reference value.

[0074] Furthermore, in the unified unit of coordinate and control loop, the control principle model of the grid-type converter satisfies formula (18)-(22). In order to achieve accurate tracking of reactive power, an integral element is introduced in the control loop, as shown in formula (23). The coordinate system transformation satisfies formula (24), the reference current of the grid-type converter satisfies formula (25)-(26), the reference current of the grid-type converter satisfies formula (27)-(28), and the reference voltage signal calculation satisfies formula (29)-(31).

[0075] (18)

[0076] (19)

[0077] (20)

[0078] (twenty one)

[0079] (twenty two)

[0080] (twenty three)

[0081] (twenty four)

[0082] (25)

[0083] (26)

[0084] (27)

[0085] (28)

[0086] (29)

[0087] (30)

[0088] In the formula: It is the equivalent moment of inertia; The system angular velocity; Rated angular velocity; This is the equivalent input power; Electromagnetic power; The damping coefficient; Rated active power; This is the active power droop coefficient; Rated voltage; This is the reactive power droop coefficient; Rated reactive power; For rotation angle; in GFM mode, it is... In GFL mode, ; , These are the proportional-integral adjustment parameters; For network mode Shaft reference current value; For network mode Shaft reference current value; For network construction mode Shaft reference current value; The internal potential of the system; Terminal voltage; Terminal current; For network construction mode Shaft reference current value; To generate the reference PWM signal Shaft voltage value; To generate the reference PWM signal Shaft voltage value; , Each represents the current time. Actual output value of shaft current; , , The three-phase voltage values ​​are used to generate the reference PWM signal.

[0089] Furthermore, the hybrid system integrated control module enables the system to operate stably under strong grid, weak grid and transitional operating conditions, and reduces the total harmonic distortion rate of grid-connected current to below 0.85% through the long time domain model prediction control module, and achieves no impact on current, voltage and frequency when switching between grid-type and grid-connected converter modes through the disturbance-free switching control module.

[0090] Furthermore, the long-term model prediction control module and the disturbance-free switching control module are functionally independent and complementary, jointly improving the overall performance of the hybrid system and adapting to the operation requirements of the power system under the condition of high proportion of new energy access.

[0091] The advantages and beneficial effects of this invention are as follows: This invention achieves coordinated control and stable operation of a hybrid grid-connected and grid-connected converter system, improving dynamic response speed, power quality, and operational flexibility while maintaining system stability. This invention replaces the traditional PI current inner loop with long-term domain model predictive control, reducing the total harmonic distortion (THD) of the grid-connected current from 5.40% to 0.85%, significantly reducing frequency fluctuations. Furthermore, it achieves a seamless transition during GFM / GFL mode switching, with parameters such as current, voltage, and power all switching smoothly, significantly enhancing the stable operation capability of the power system under conditions of high renewable energy integration. Attached Figure Description

[0092] Figure 1 This is a flowchart of the present invention;

[0093] Figure 2 Here are the control framework diagrams for the system; (a) is the overall control framework diagram for the grid-type converter, (b) is the overall control framework diagram for the grid-connected converter, and (c) is the overall control framework diagram for the hybrid system.

[0094] Figure 3 The diagram shows the effect of long-time domain model predictive control; (a) is the d-axis tracking reference current diagram of long-time domain model predictive control, (b) and (c) are the grid-connected frequency waveform diagrams of traditional PI control and long-time domain model predictive control, and (d) and (e) are the grid-connected current spectrum analysis diagrams of traditional PI control / long-time domain model predictive control.

[0095] Figure 4The diagram shows the effect of the grid-connected system's seamless switching control. Among them, (a) is the power waveform of the seamless switching system without switching impedance, (b) and (c) are the output current and voltage waveforms of the seamless switching system without switching impedance, (d) is the power waveform of the seamless switching system with switching impedance, (e) and (f) are the output current and voltage waveforms of the seamless switching system with switching impedance, and (g) is the frequency waveform of the grid connection point with seamless switching impedance. Detailed Implementation

[0096] To make the objectives, technical solutions, and advantages of the present invention clearer, the present invention will be further described in detail below with reference to the embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0097] Example 1:

[0098] like Figure 1 As shown, a method for long-time-domain predictive control and disturbance-free switching of a grid-connected converter in a hybrid system includes the following steps:

[0099] S1: Construct a long-time-domain model predictive current control strategy for grid-connected converters to replace the current PI inner loop in traditional virtual synchronous generator control.

[0100] A discretized predictive model for a grid-type converter is established. Based on the dynamic equation of the output current in the synchronous rotating dq coordinate system, the forward Euler method is used for discretization to obtain a discrete state-space model that can be used for predictive control.

[0101] This invention designs a long-time-domain predictive controller. Addressing the problem that traditional single-step predictive models cannot fully compensate for system delays, this invention employs a long-time-domain predictive model. By predicting the current values ​​for two future time steps and applying different voltage vectors, it effectively compensates for system delays. This method can assess the optimal operating state of the switching transistors in advance, significantly improving the current tracking divergence phenomenon caused by the long system response time.

[0102] The reference current is updated in real time using the Lagrange extrapolation method. To ensure the accuracy of predictive control, the reference current needs to be calculated in real time. This invention employs a second-order Lagrange extrapolation algorithm, which can accurately predict the changing trend of the reference current, providing an accurate tracking target for model predictive control.

[0103] The optimal voltage vector is selected by solving the value function through rolling optimization. The value function is the core of model predictive control, used to quantify the deviation between the reference current and the predicted current. By enumerating all possible voltage vectors and calculating their corresponding value function values, the voltage vector that minimizes the value function is finally selected as the optimal control quantity. This includes seven possible voltage vectors. The operating state of the switching transistor can be 0 or 1, resulting in eight possible switching state combinations. Zero vectors may overlap, thus corresponding to seven combinations.

[0104] S2: Design a disturbance-free switching control strategy between mesh-type and follow-me-type converters:

[0105] A control principle model for a grid-connected converter is established, unifying the coordinate system and current control loop in both grid-connected and grid-following converter modes. To achieve smooth switching, the coordinate systems of the two modes must first be unified. This invention transforms the terminal voltage coordinate system in the grid-following converter mode to the αβ coordinate system, achieving coordinate system unification with the grid-connected converter mode. Simultaneously, the grid-following and grid-connected converter modes share the current control loop, ensuring a smooth transition of current commands during switching.

[0106] Grid-type control calculates and locks the switching angle in real time based on the power synchronization characteristics of a simulated traditional synchronous generator. In grid-type converter mode, the required angle for grid-type converter mode is calculated in real time using the power synchronization loop. The angle is locked during switching to ensure consistency and continuity of control variables before and after switching. Since the switching occurs in grid-connected mode, the active power command controlled by the grid-connected converter and the grid-connected converter is... When set to the same value, the angular velocity output by the power synchronization loop and the angular velocity output by the grid-connected control unit remain consistent in steady state, thus enabling smooth switching of the grid synchronization section.

[0107] When switching from grid-connected converter control to integrated grid converter control, the grid synchronization component should switch from a power synchronization loop to a phase-locked loop. During the switching process, the output of the power synchronization loop integrator... Set the initial value for the PLL unit integrator, and simultaneously switch the grid synchronization section from grid-connected mode to grid-following mode. When switching from grid-following converter control to grid-connected converter control, the grid synchronization section should switch from the PLL unit to a power synchronization loop. At this time, the output of the PLL unit integrator needs to be set to the initial value. Set the initial value for the power synchronization loop integrator and switch the switch from grid-following mode to grid-connecting mode.

[0108] During the switching process, the reference voltage value is dynamically adjusted based on the difference between the internal potential and the terminal voltage. The reference voltage value is modulated by SPWM to obtain the PWM duty cycle, and then the PWM control signal is dynamically adjusted to ensure that the key variables remain continuous at the moment of switching and avoid sudden changes.

[0109] S3: Integrate the long-time-domain model predictive control strategy and the disturbance-free switching strategy into the hybrid system control framework to achieve stable operation and smooth mode switching of the system under strong and weak power grid conditions.

[0110] In a hybrid system, long-time-domain model predictive current control is implemented in the grid-type converter to establish a complete current predictive control system. The aim is to achieve rapid and accurate tracking of the reference current and elimination of harmonic distortion of the grid-connected current. A disturbance-free switching control mechanism is also configured to ensure that no impact on parameters such as current, voltage and frequency is generated during the mode switching process.

[0111] In a hybrid system, long-time-domain model predictive control focuses on improving the dynamic performance and steady-state accuracy of the grid-type converter, while disturbance-free switching control is responsible for ensuring the smoothness of system operation mode transitions. The two are functionally independent and complementary, working together to improve the overall performance of the hybrid system.

[0112] To meet the operational needs of hybrid systems under different grid conditions, the system can maintain stable operation under strong grid, weak grid and transitional conditions, and achieve smooth transition during mode switching, significantly improving the operational flexibility and stability of the power system under conditions of high proportion of new energy access.

[0113] Furthermore, in step S1, the predictive control model of the grid-type converter is shown in formulas (1)-(6), wherein the relationship between the output voltage vector V and the switching signal S is shown in formula (4), the forward Euler method is shown in formula (7), which discretizes the model, the discrete state space model is shown in formulas (8)-(9), the single-step prediction model is shown in formulas (10)-(11), the long-time domain prediction model is shown in formulas (12)-(13), the value function is shown in formula (14), the second-order Lagrange extrapolation algorithm is shown in formula (15), and the derivation of the reference current values ​​for the next two sampling times is shown in formulas (16)-(17).

[0114] (1)

[0115] (2)

[0116] (3)

[0117] (4)

[0118] (5)

[0119] (6)

[0120] (7)

[0121] (8)

[0122] (9)

[0123] (10)

[0124] (11)

[0125] (12)

[0126] (13)

[0127] (14)

[0128] (15)

[0129] (16)

[0130] (17)

[0131] In the formula: These are the switching signals for a three-phase inverter, representing each switching state. , , It can be 0 or 1; It is a DC voltage; Let be the output voltage vector, where , , The value is Or it could be 0; For grid-connected current Axial components; For grid-connected current Axial components; For converter Shaft output voltage; For converter Shaft output voltage; For grid voltage Axial components; For grid voltage Axial components; , These are the filter inductor and resistor, respectively. The system angular velocity; This is the sampling period of the system, i.e., the time of a single step. For the first Each sampling period Predicted shaft current value; For the first Each sampling period Predicted shaft current value; For the first Each sampling period Shaft current reference value; For the first Each sampling period Shaft current reference value; For the first Each sampling period Predicted shaft current value; For the first Each sampling period Predicted shaft current value; For the first Each sampling period Shaft current reference value; For the first Each sampling period Shaft current reference value.

[0132] Furthermore, in step S2, a control principle model for the grid-type converter is established, as shown in formulas (18)-(22). To achieve accurate tracking of reactive power, an integral element is introduced into the control loop, as shown in formula (23). The coordinate system transformation is shown in formula (24). The reference current of the grid-type converter is shown in formulas (25)-(26). The reference current of the grid-type converter is shown in formulas (27)-(28). The reference voltage signal is calculated as shown in formulas (29)-(31).

[0133] (18)

[0134] (19)

[0135] (20)

[0136] (twenty one)

[0137] (twenty two)

[0138] (twenty three)

[0139] (twenty four)

[0140] (25)

[0141] (26)

[0142] (27)

[0143] (28)

[0144] (29)

[0145] (30)

[0146] (31)

[0147] In the formula: It is the equivalent moment of inertia; The system angular velocity; Rated angular velocity; This is the equivalent input power; Electromagnetic power; The damping coefficient; Rated active power; This is the active power droop coefficient; Rated voltage; This is the reactive power droop coefficient; Rated reactive power; For rotation angle; in GFM mode, it is... In GFL mode, ; , These are the proportional-integral adjustment parameters; For network mode Shaft reference current value; For network mode Shaft reference current value; For network construction mode Shaft reference current value; The internal potential of the system; Terminal voltage; Terminal current; For network construction mode Shaft reference current value; To generate the reference PWM signal Shaft voltage value; To generate the reference PWM signal Shaft voltage value; , Each represents the current time. Actual output value of shaft current; , , The three-phase voltage values ​​are used to generate the reference PWM signal.

[0148] The advantages and beneficial effects of this invention are as follows: This invention achieves coordinated control and stable operation of a hybrid grid-connected and grid-connected converter system, improving dynamic response speed, power quality, and operational flexibility while maintaining system stability. This invention replaces the traditional PI current inner loop with long-term domain model predictive control, reducing the total harmonic distortion (THD) of the grid-connected current from 5.40% to 0.85%, significantly reducing frequency fluctuations. Furthermore, it achieves a seamless transition during GFM / GFL mode switching, with parameters such as current, voltage, and power all switching smoothly, significantly enhancing the stable operation capability of the power system under conditions of high renewable energy integration.

[0149] Example 2

[0150] like Figure 3 As shown, a frequency drop of 0.1Hz is set in the system from 0.5s to 1s to verify the effectiveness of the proposed algorithm. From Figure 3 As can be seen from (a), after long-time domain model predictive control, the current fluctuation phenomenon is small when the frequency changes abruptly at 0.5s and 1s after system startup. Furthermore, the current tracking speed is fast, and at 0.2s, it is essentially error-free and close to the reference current. Figure 3 As can be seen from (b) and (c), under traditional PI control, the frequency of the grid connection point cannot quickly recover to a stable state during frequency mutations, and obvious fluctuations occur after the overshoot peak. However, the algorithm proposed in this paper can quickly stabilize the frequency during frequency mutations, verifying the effectiveness of the proposed algorithm for frequency regulation of hybrid systems. Figure 3 As can be seen from (d) and (e), the traditional PI control algorithm has a slow response speed and lacks strong adaptability when the frequency changes abruptly, resulting in severe distortion, with a total harmonic distortion rate as high as 5.40%. In contrast, the algorithm proposed in this paper can quickly track the reference current value when the frequency changes abruptly, thereby reducing the current distortion phenomenon and lowering the total harmonic distortion rate to 0.85%. This verifies that the algorithm proposed in this paper can effectively improve the current distortion phenomenon at the grid connection and improve the power quality of the power grid.

[0151] like Figure 4 As shown, switching signals are set at 0.5s and 1s respectively to achieve system switching from GFM to GFL to GFM. Figure 4As can be seen from (a), (b), and (c), when the impedance is not switched, the control system achieves a disturbance-free switching 0.5s after system startup. The changes in active and reactive power in the system are small and quickly recover to the given values. Furthermore, the output voltage and current do not fluctuate significantly during the GFM / GFL mode switching, indicating that the proposed algorithm has minimal impact on the system's operating state and can effectively achieve disturbance-free switching between GFM and GFL. To verify the robustness of the proposed algorithm, the system impedance is switched at 0.5s and 1s respectively to achieve mutual conversion between strong and weak power grids, and the switching from GFM to GFL to GFM under this condition is verified. Figure 4 As shown in (d), (e), and (f), when the impedance is switched, the active and reactive power of the system basically do not fluctuate after 0.5s of system startup, achieving a disturbance-free switching. At 1s, the power fluctuates slightly but quickly recovers to the given value. The output voltage and current of the system fluctuate only slightly when the impedance is switched, indicating that the algorithm designed in this paper can adapt well to the switching between strong and weak power grids and achieve disturbance-free switching between GFM and GFL modes. Figure 4 As can be seen from (g), when impedance switching occurs 0.5s and 1s after system startup, the GFM / GFL switching strategy can effectively control the frequency. The frequency at the grid connection point only fluctuates slightly during impedance switching and quickly recovers to 50Hz. This indicates that the disturbance-free switching method proposed in this paper can be well implemented without significant frequency fluctuations, and the system parameters transition smoothly, improving the flexibility of power system operation.

Claims

1. A method for long-time-domain predictive control and disturbance-free switching of a grid-connected converter in a hybrid system, characterized in that, Includes the following steps: S1. Construct a long-time-domain model predictive current control strategy for grid-type converters to replace the current PI inner loop in traditional virtual synchronous generator control: S11: Based on the dynamic equation of the output current in the synchronously rotating dq coordinate system, the discrete state-space model is obtained by discretization using the forward Euler method. S12: Design a long-time-domain prediction model to compensate for system delay by predicting the current values ​​for the next two time steps and applying different voltage vectors. S13: The reference current is updated in real time using the second-order Lagrange extrapolation method to provide an accurate tracking target for predictive control; S14: Solve the value function through rolling optimization, enumerate all possible voltage vectors and select the voltage vector that minimizes the value function as the optimal control quantity; Among them, the predictive control model of the grid-type converter is shown in formulas (1)-(6), the forward Euler method is shown in formula (7), the model is discretized, the discrete state space model is shown in formulas (8)-(9), the single-step prediction model is shown in formulas (10)-(11), the long-time domain prediction model is shown in formulas (12)-(13), the value function is shown in formula (14), the second-order Lagrange extrapolation algorithm is shown in formula (15), and the derivation of the reference current values ​​for the next two sampling times is shown in formulas (16)-(17). （1） （2） （3） （4） （5） （6） （7） （8） （9） （10） （11） （12） （13） （14） （15） （16） （17） In the formula: These are the switching signals for a three-phase inverter, representing each switching state. , , It can be 0 or 1; It is a DC voltage; Let be the output voltage vector, where , , The value is Or it could be 0; For grid-connected current Axial components; For grid-connected current Axial components; For converter Shaft output voltage; For converter Shaft output voltage; For grid voltage Axial components; For grid voltage Axial components; , These are the filter inductor and resistor, respectively. The system angular velocity; This is the sampling period of the system, i.e., the time of a single step. For the first Each sampling period Predicted shaft current value; For the first Each sampling period Predicted shaft current value; For the first Each sampling period Shaft current reference value; For the first Each sampling period Shaft current reference value; For the first Each sampling period Predicted shaft current value; For the first Each sampling period Predicted shaft current value; For the first Each sampling period Shaft current reference value; For the first Each sampling period Shaft current reference value; S2. Design a disturbance-free switching control strategy between grid-type and follow-grid-type converters: S21. Establish a control principle model for a grid-type converter, transform the terminal voltage coordinate system in the grid-type converter mode to the αβ coordinate system, realize the same coordinate system as the grid-type converter mode, and share the current control loop between the two modes. S22. In grid-type converter mode, the required angle for grid-type converter mode is calculated in real time using the power synchronization loop. The angle is locked during switching because the switching occurs in grid-connected mode, and the active power command controlled by the grid-connected converter and the grid-connected converter is... When set to the same value, the angular velocity output by the power synchronization loop and the angular velocity output by the grid-type control unit remain consistent under steady state. S23. When switching from a grid-type converter to a follow-grid type converter, the grid synchronization part changes from a power synchronization loop to a phase-locked loop. Set the initial value for the phase-locked loop integrator; when switching from a grid-type to a grid-type converter, the grid synchronization part changes from a phase-locked loop to a power synchronization loop. Set the initial value for the power synchronization loop integrator; S24. During the switching process, the reference voltage value is dynamically adjusted based on the difference between the internal potential and the terminal voltage. The reference voltage value is modulated by SPWM to obtain the PWM duty cycle, and then the PWM control signal is dynamically adjusted to keep the variables continuous. Among them, the established control principle model of the grid-type converter is shown in formulas (18)-(22). An integral element is introduced into the control loop as shown in formula (23). The coordinate system transformation is shown in formula (24). The reference current of the grid-type converter is shown in formulas (25)-(26). The reference current of the grid-type converter is shown in formulas (27)-(28). The reference voltage signal is calculated as shown in formulas (29)-(31). （18） （19） （20） （21） （22） （23） （24） （25） （26） （27） （28） （29） （30） （31） In the formula: It is the equivalent moment of inertia; The system angular velocity; Rated angular velocity; This is the equivalent input power; Electromagnetic power; The damping coefficient; Rated active power; This is the active power droop coefficient; Rated voltage; This is the reactive power droop coefficient; Rated reactive power; This is the rotation angle; in GFM mode, it is... In GFL mode, ; , These are the proportional-integral adjustment parameters; For network mode Shaft reference current value; For network mode Shaft reference current value; For network construction mode Shaft reference current value; The internal potential of the system; Terminal voltage; Terminal current; For network construction mode Shaft reference current value; To generate the reference PWM signal Shaft voltage value; To generate the reference PWM signal Shaft voltage value; , Each represents the current time. Actual output value of shaft current; , , To generate the three-phase voltage values ​​for the reference PWM signal; S3. Integrate long-time-domain model predictive control strategy and disturbance-free switching strategy into the hybrid system control framework. Implement long-time-domain model predictive current control in grid-type converters and configure disturbance-free switching control mechanism to achieve stable operation and smooth mode switching of the system under strong grid, weak grid and transition conditions.

2. A long-time-domain predictive control and disturbance-free switching system for a hybrid grid-connected converter, characterized in that, include: Long-term model predictive control module, disturbance-free switching control module, and hybrid system integrated control module; The long-time-domain model predictive control module is configured to construct a long-time-domain model predictive current control strategy for the grid-type converter, replacing the current PI inner loop in the traditional virtual synchronous generator control. The long-time-domain model predictive control module includes a discretization modeling unit, a long-time-domain prediction unit, a reference current update unit, and an optimization solution unit. The discretization modeling unit, based on the output current dynamic equation in the synchronous rotating dq coordinate system, uses the forward Euler method for discretization to establish a discrete state-space model of the grid-type converter. The long-time-domain prediction unit designs a long-time-domain prediction model, compensating for system delay by predicting the current values ​​for two future time steps and applying different voltage vectors. The reference current update unit uses a second-order Lagrange extrapolation algorithm to update the reference current in real time, providing a quasi-current for predictive control. Accurately track the target; the optimization solution unit, through rolling optimization solution of the value function, enumerates all possible voltage vectors and selects the voltage vector that minimizes the value function as the optimal control quantity; in the discretization modeling unit, the network converter predictive control model satisfies formula (1)-(6), the forward Euler method satisfies formula (7), and the discrete state space model satisfies formula (8)-(9); in the long time domain prediction unit, the single-step prediction model satisfies formula (10)-(11), and the long time domain prediction model satisfies formula (12)-(13); in the optimization solution unit, the value function satisfies formula (14); in the reference current update unit, the second-order Lagrange extrapolation algorithm satisfies formula (15), and the derivation of the reference current value for the next two sampling times satisfies formula (16)-(17); （1） （2） （3） （4） （5） （6） （7） （8） （9） （10） （11） （12） （13） （14） （15） （16） （17） In the formula: These are the switching signals for a three-phase inverter, representing each switching state. , , It can be 0 or 1; It is a DC voltage; Let be the output voltage vector, where , , The value is Or it could be 0; For grid-connected current Axial components; For grid-connected current Axial components; For converter Shaft output voltage; For converter Shaft output voltage; For grid voltage Axial components; For grid voltage Axial components; , These are the filter inductor and resistor, respectively. The system angular velocity; This is the sampling period of the system, i.e., the time of a single step. For the first Each sampling period Predicted shaft current value; For the first Each sampling period Predicted shaft current value; For the first Each sampling period Shaft current reference value; For the first Each sampling period Shaft current reference value; For the first Each sampling period Predicted shaft current value; For the first Each sampling period Predicted shaft current value; For the first Each sampling period Shaft current reference value; For the first Each sampling period Shaft current reference value; The disturbance-free switching control module is configured to achieve disturbance-free switching between grid-type and follow-grid converters. It includes a coordinate and control loop unification unit, a switching angle calculation and locking unit, a synchronous switching unit, and a variable coordination control unit. The coordinate and control loop unification unit establishes a control principle model for the grid-type converter, transforming the terminal voltage coordinate system in the follow-grid converter mode to the αβ coordinate system, achieving coordinate system unification with the grid-type converter mode, and enabling both modes to share the current control loop. The switching angle calculation and locking unit, in the grid-type converter mode, uses the power synchronization loop to calculate the required angle for the follow-grid converter mode in real time. And lock the angle during switching; the synchronous switching unit, when switching from a grid configuration to a follow grid configuration, controls the grid synchronization part to switch from a power synchronization loop to a phase-locked loop, and... Set the initial value for the phase-locked loop integrator; when switching from a follow-the-grid type to a multi-grid type, the control grid synchronization part switches from a phase-locked loop to a power synchronization loop, and... Set as the initial value of the power synchronization loop integrator; the variable coordination control unit coordinates the internal potential, terminal voltage, and terminal current during the switching process, calculates the reference PWM control signal, and ensures that the key variables are continuous without sudden changes; in the coordinate and control loop unified unit, the grid-type converter control principle model satisfies formula (18)-(22), and an integral element is introduced in the control loop unified unit, as shown in formula (23). The coordinate system transformation satisfies formula (24), the grid-type converter reference current satisfies formula (25)-(26), the grid-type converter reference current satisfies formula (27)-(28), and the reference voltage signal calculation satisfies formula (29)-(31); （18） （19） （20） （21） （22） （23） （24） （25） （26） （27） （28） （29） （30） （31） In the formula: It is the equivalent moment of inertia; The system angular velocity; Rated angular velocity; This is the equivalent input power; Electromagnetic power; The damping coefficient; Rated active power; This is the active power droop coefficient; Rated voltage; This is the reactive power droop coefficient; Rated reactive power; This is the rotation angle; in GFM mode, it is... In GFL mode, ; , These are the proportional-integral adjustment parameters; For network mode Shaft reference current value; For network mode Shaft reference current value; For network construction mode Shaft reference current value; The internal potential of the system; Terminal voltage; Terminal current; For network construction mode Shaft reference current value; To generate the reference PWM signal Shaft voltage value; To generate the reference PWM signal Shaft voltage value; , Each represents the current time. Actual output value of shaft current; , , To generate the three-phase voltage values ​​for the reference PWM signal; The hybrid system integrated control module is configured to integrate long-time domain model predictive control (LTVMA) and seamless switching strategies into the hybrid system control framework. This enables stable operation and smooth mode switching of the system under strong and weak grid conditions. The LTVMA focuses on improving the dynamic performance and steady-state accuracy of grid-connected converters, while the seamless switching control ensures the smoothness of system operation mode transitions. The hybrid system integrated control module enables stable operation of the system under strong, weak, and transitional grid conditions. The LTVMA reduces the total harmonic distortion (THD) of the grid-connected current to below 0.85%, and the seamless switching control ensures no impact on current, voltage, or frequency during mode switching between grid-connected and grid-connected converters. The LTVMA and seamless switching control modules are functionally independent yet complementary, jointly improving the overall performance of the hybrid system and adapting to the operational needs of the power system under conditions of high renewable energy integration.

Images