A grid-connected converter model predictive control method and control system

Abstract

The application provides a grid-connected converter model predictive control method and a control system, and belongs to the technical field of power electronics. The method determines the voltage at the converter side according to a defined switching period, establishes a cost function of switching of a switching device of the control converter, obtains an optimal problem expression of the next switching control period when the cost function is minimum according to the obtained circuit parameter values of the converter output end in the current switching control period, obtains the switching state of the switching tube of the converter in the next switching control period according to the optimal problem expression, and generates a trigger signal to control the opening and closing of the switching device of the converter according to the switching state. The control function of the application embodiment retains the dynamic characteristics of the LCL grid-connected converter model and completes the model-based predictive control thereof under the premise of the first order, increases the dynamic characteristics of the system, and can realize the optimization of multiple targets.

Description

Technical Field

[0001] This invention relates to the field of power electronics technology, and in particular to a model predictive control method and control system for grid-connected converters. Background Technology

[0002] Generally speaking, grid-connected converters have a filter at the output end to suppress the output harmonics of the converter. LCL filters are widely used in grid-connected converter products because of their small size and high-frequency harmonic attenuation capabilities.

[0003] Traditionally, proportional-integral (PI), proportional-resonant (PR), and repetitive control methods are used to control grid-connected converters. However, these methods struggle to achieve multi-objective optimization. Model predictive control (MMC) offers advantages such as good dynamic performance and the ability to perform multi-objective optimization, providing strong support for improving and optimizing the performance of grid-connected converters.

[0004] With the continuous development of control chips, their performance has gradually improved without significant cost changes, making it increasingly appropriate to apply model predictive control (MMC) to grid-connected converter products. MMC requires predicting the physical quantities for the next control cycle based on the converter's mathematical model to select the converter's switching state. However, a grid-connected converter using an LCL filter is a high-order system, making the proper handling of its model and control a technical challenge in engineering. Therefore, simplified model methods are often used, such as reducing the LCL filter to an L-filter for control, or only considering the low-frequency characteristics of the model during control. However, this affects the consideration of the model's dynamic characteristics. Summary of the Invention

[0005] In view of this, the purpose of the present invention is to provide a model predictive control method and control system for grid-connected converters, which aims to solve the technical problems of existing model predictions affecting the dynamic characteristics of the model and avoiding the use of high-order equations.

[0006] The technical solution adopted by the present invention to solve the above-mentioned technical problems is as follows:

[0007] A first aspect of the present invention provides a model predictive control method for a grid-connected converter, comprising the following steps:

[0008] Define the switching control cycle of the converter;

[0009] The voltage on the converter side is determined according to the defined switching cycle in order to achieve current tracking and control the rate of change of current. A cost function for controlling the switching of the switching devices in the converter is established.

[0010] Obtain the circuit parameter values ​​of the converter output terminal collected during the current switching control cycle;

[0011] Based on the obtained circuit parameter values ​​at the converter output during the current switching control cycle, the optimal problem expression for the next switching control cycle when the cost function is minimized is obtained.

[0012] Based on the optimal problem expression, the switching state of the converter's switching transistors in the next switching control cycle is calculated, and a trigger signal is generated based on the switching state to control the opening and closing of the converter's switching devices.

[0013] In some embodiments, defining the switching control cycle of the converter includes: defining S as a switching state, S=1, and the converter-side voltage as V. busp； When S = 0, the converter-side voltage is 0; when S = -1, the converter-side voltage is -V. busn The interval between adjacent switch conduction actions is defined as the rising cycle, and the interval between adjacent switch disconnection actions is defined as the falling cycle. The arithmetic mean of the rising cycle and the falling cycle is defined as the switch control cycle.

[0014] In some embodiments, the cost function includes a given current tracking term, a control current rate of change term, a cost coefficient term, and a switching frequency optimization term; the control current rate of change term has the function of suppressing resonance, the cost coefficient is adjusted according to specific engineering requirements, and the larger the coefficient, the stronger the suppression effect on resonance; the switching frequency optimization term is used to optimize the average switching frequency, and the larger the cost coefficient, the lower the average switching frequency of the converter.

[0015] In some embodiments, the cost function for controlling the switching devices of the converter is as follows:

[0016]

[0017] in

[0018] This is the reference current for the current control cycle on the converter side;

[0019] This is the current estimate for the current cycle on the converter side.

[0020] This is an estimated value of the current during the current control cycle on the grid side;

[0021] This represents the change in current on the converter side during the current control cycle.

[0022] k j1 k j2 k j3 These are the cost coefficients, where k j1 k j2 All are less than 1;

[0023] Sk Indicates the on / off state of the current control cycle;

[0024] S k-1 It indicates the switching state of the previous control cycle.

[0025] In one embodiment, a method for obtaining the optimal problem expression for the next switching control cycle corresponding to the minimum cost function based on the obtained circuit parameter values ​​at the converter output during the current switching control cycle includes the following steps:

[0026] Based on the circuit parameter values ​​of the converter at the current acquisition time, calculate the output current value of the DC capacitor of the converter at the end of the current control cycle.

[0027] Based on the DC capacitor voltage difference and output current value of the converter at the end of the current control cycle, the values ​​of all output voltage states of the converter in the next control cycle are rolled over and substituted to predict and calculate the capacitor output current of the converter at the end of the next control cycle.

[0028] Based on all the output voltage states of the converter in the next control cycle, calculate the switching state corresponding to each output voltage state in the next control cycle; based on the converter output voltage collected at the previous moment and the current moment, calculate the switching state corresponding to the output voltage state at the previous moment and the current moment.

[0029] Based on the predicted output current of the converter and the difference between the DC capacitor current at the end of the next control cycle, the switching state corresponding to each output voltage state in the next control cycle, and the switching state corresponding to the output voltage state at the previous and current times, calculate the value of the cost function for each output voltage state in the next control cycle.

[0030] The converter output voltage state corresponding to the minimum cost function value is set as the optimal problem expression for the next switching control cycle of the converter.

[0031] In one embodiment, the optimal problem expression for obtaining the minimum cost function in the next switching control cycle is:

[0032]

[0033] Wherein, min(J k Minimum of the cost function, S k Indicates the on / off state of the current control cycle;

[0034] S k-1 It indicates the switching state of the previous control cycle.

[0035] In one embodiment, the calculation of the current value of the output DC capacitor of the converter at the end of the current control cycle, based on the circuit parameter values ​​of the converter at the current acquisition time, includes the following formula:

[0036] i con =k ref i ref +(1-k ref )i g (6)

[0037]

[0038] This is the reference value for the current in the current control cycle;

[0039] This is the capacitor current value for the current control cycle;

[0040] k ref To introduce a given rate of change factor, its range is (0,1];

[0041] i ref This is the reference current on the grid side;

[0042] i g This refers to the grid-side current.

[0043] This represents the voltage of the filter capacitor during the current control cycle.

[0044] This is the voltage of the filter capacitor in the previous control cycle;

[0045] T s To control and sample the period;

[0046] C is the AC filter capacitor;

[0047] Based on the calculated DC capacitor voltage difference and output current value of the converter at the end of the current control cycle, the values ​​of all output voltage states of the converter in the next control cycle are substituted in a rolling manner to predict and calculate the capacitor output current of the converter at the end of the next control cycle. The discretized formula is as follows:

[0048]

[0049]

[0050] The capacitor current value for the current control cycle can be obtained from formulas (7), (8), and (9). for:

[0051]

[0052] This is the reference current on the grid side during the current control cycle;

[0053] This is the reference current for the grid side in the previous control cycle;

[0054] This is the reference current on the grid side in the previous control cycle;

[0055] The voltage on the grid side during the current control cycle;

[0056] L g This represents the current in the filter inductor;

[0057] This represents the voltage of the filter capacitor during the current control cycle.

[0058] This is the voltage of the filter capacitor in the previous control cycle;

[0059] T s To control and sample the period;

[0060] C is the AC filter capacitor.

[0061] In one embodiment, the estimated value of the grid-side current The expression is:

[0062]

[0063] Among them, T s Control and adoption cycle;

[0064] L g This is the inductance value of the filter inductor on the grid side;

[0065] The voltage on the grid side during the current control cycle;

[0066] The voltage of the filter capacitor in the current control cycle;

[0067] This represents the current on the grid side during the previous control cycle.

[0068] A second aspect of the present invention also provides a grid-connected converter model predictive control system, the system comprising a definition module, a cost function establishment module, an acquisition module, a calculation module, and a control output module that are interconnected.

[0069] The definition module is configured to define the switching control cycle of the converter;

[0070] The cost function establishment module is configured to determine the voltage on the converter side according to the defined switching cycle, with the goal of current tracking and controlling the balance of the rate of change of current, and to establish a cost function for controlling the switching of the switching devices of the converter.

[0071] The acquisition module is configured to acquire circuit parameter values ​​of the converter output terminal collected during the current switching control cycle.

[0072] The calculation module is configured to obtain the optimal problem expression for the next switching control cycle when the cost function is minimized, based on the obtained circuit parameter values ​​of the converter output in the current switching control cycle.

[0073] The control output module is configured to calculate the switching state of the converter's switching transistors in the next switching control cycle based on the optimal problem expression, and generate a trigger signal to control the opening and closing of the converter's switching devices based on the switching state.

[0074] In some embodiments, the calculation module includes a capacitor output current value calculation unit, a switch state acquisition unit, a cost function calculation unit, and an optimal calculation unit;

[0075] The capacitor output current calculation unit is used to calculate the output current value of the DC capacitor of the converter at the end of the current control cycle based on the circuit parameter values ​​of the converter at the current acquisition time.

[0076] The switching state acquisition unit is used to calculate the switching state corresponding to each output voltage state in the next control cycle based on all output voltage states of the converter in the next control cycle; and to calculate the switching state corresponding to the output voltage states in the previous and current times based on the converter output voltages collected in the previous and current times.

[0077] The cost function calculation unit is used to calculate the value of the cost function for each output voltage state in the next control cycle based on the difference between the output current of the converter and the DC capacitor current at the end of the predicted next control cycle, the switching state corresponding to each output voltage state in the next control cycle, and the switching state corresponding to the output voltage state at the previous moment and the current moment.

[0078] The optimal calculation unit is used to set the converter output voltage state corresponding to the minimum cost function value as the optimal problem expression for the next switching control cycle of the converter.

[0079] The grid-connected converter model predictive control method and control system provided in this embodiment of the invention retain the dynamic characteristics of the LCL grid-connected converter model and complete its model-based predictive control under the premise that the control function of the grid-connected converter model is first-order, thereby increasing the dynamic characteristics of the system and achieving the optimization of multiple objectives. Attached Figure Description

[0080] Figure 1 This is a topology circuit diagram of a grid-connected converter according to an embodiment of the present invention;

[0081] Figure 2 This is a flowchart of an embodiment of a model predictive control method for a grid-connected converter provided by the present invention;

[0082] Figure 3 This is a single-phase equivalent model circuit diagram of a three-level grid-connected converter with an LCL filter according to an embodiment of the present invention;

[0083] Figure 4 This is a flowchart of another embodiment of the model predictive control method for grid-connected converters according to an embodiment of the present invention;

[0084] Figure 5 This is a structural block diagram of an embodiment of a grid-connected converter model predictive control system according to an embodiment of the present invention;

[0085] Figure 6 This is a structural block diagram of another embodiment of a grid-connected converter model predictive control system according to an embodiment of the present invention. Detailed Implementation

[0086] To make the technical problems, solutions, and beneficial effects of this invention clearer and more understandable, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0087] Example 1:

[0088] This application provides a model predictive control method for grid-connected converters, applied in a model predictive control system for grid-connected converters. Please refer to [link to relevant documentation]. Figure 1 This is the topology diagram of a grid-connected converter. An LCL filter serves as its filtering interface with the power grid. ga L gb ,L gc These are three-phase filter inductors on the grid side, all with the same inductance value of L. g C a C b C c These are three-phase AC filter capacitors, all with the same capacitance value (C, L).inva L invb ,L invc These are the filter inductors on the converter side, all with the same inductance value of L. inv C DC This is a DC bus filter capacitor. ga u gb u gc These are the sampled voltages of phases A, B, and C of the power grid, with the voltage reference points being... Figure 1 The method described in this application uses a single-phase converter model for system analysis, modeling, and control due to the symmetry of the topology.

[0089] Please see Figure 2 and Figure 3 The grid-connected converter model predictive control method includes the following steps:

[0090] S101. Define the switching control cycle of the converter;

[0091] Specifically, please refer to Figure 3 This is a single-phase equivalent model circuit diagram of a three-level grid-connected converter with an LCL filter. S is defined as the switching state, S=1, and the converter-side voltage is V. busp When S = 0, the converter-side voltage is 0; when S = -1, the converter-side voltage is -V. busn The interval between adjacent switch-on actions is defined as the rising cycle, and the interval between adjacent switch-off actions is defined as the falling cycle. The arithmetic mean of the rising cycle and the falling cycle is defined as the switch control cycle.

[0092] Let the reference current on the grid side be i ref The control and sampling period is T. s The voltage on the converter side is u inv The grid-side voltage is u g The grid-side current is i g The inverter-side current is i inv The capacitor current is i c The current value of each circuit parameter is marked with a superscript k, the value of the previous control cycle is represented by a superscript k-1, the value of the next control cycle is represented by a superscript k+1, and so on.

[0093] S102. Determine the voltage on the converter side according to the defined switching control cycle in order to achieve current tracking and control the rate of change of current, and establish the cost function for controlling the switching of the switching devices of the converter.

[0094] S103. Obtain the circuit parameter values ​​of the converter output terminal collected during the current switching control cycle;

[0095] Specifically, the circuit parameter values ​​include a grid-side reference current of i.ref The control and sampling period is T. s The voltage on the converter side is u inv The grid-side voltage is u g The grid-side current is i g The inverter-side current is i inv The capacitor current is i c .

[0096] S104. Based on the obtained circuit parameter values ​​of the converter output terminal in the current switching control cycle, obtain the optimal problem expression for the next switching control cycle when the cost function is minimized.

[0097] Specifically, please refer to Figure 4 It includes the following steps:

[0098] S1041. Based on the circuit parameter values ​​of the converter at the current acquisition time, calculate the output current value of the DC capacitor of the converter at the end of the current control cycle.

[0099] Specifically, according to Figure 3 From the circuit described above, it can be seen that:

[0100]

[0101]

[0102] i c =i g -i inv (3)

[0103]

[0104] Equations (1)-(4) are the basic equations of the system shown in Figure (1). In this invention, u... g i g i inv V busp V busn For quantities that can be directly measured, i in the frequency domain can be obtained from formulas (1)-(4). g expression:

[0105]

[0106] Wherein, the converter-side voltage is u inv The grid-side voltage is u g The grid-side current is i g The inverter-side current is i inv The capacitor current is i c L inv Here, s represents the filter inductance on the converter side, and s denotes the Laplace operator.

[0107] To control the rate of change of the grid-side current, this invention introduces a given rate of change factor k. ref Define the current control reference value i con Its expression is:

[0108] i con =k ref i ref +(1-k ref )i g (6)

[0109] Where k ref The value range is (0,1).

[0110] Discretize equation (4) to obtain the capacitor current value for the current control cycle. for:

[0111]

[0112] The meanings of each letter in the above formula are as follows:

[0113] This is the reference value for the current in the current control cycle;

[0114] This is the capacitor current value for the current control cycle;

[0115] k ref To introduce a given rate of change factor, its range is (0,1];

[0116] i ref This is the reference current on the grid side;

[0117] i g This refers to the grid-side current.

[0118] This represents the voltage of the filter capacitor during the current control cycle.

[0119] This is the voltage of the filter capacitor in the previous control cycle;

[0120] T s To control and sample the period;

[0121] C is the AC filter capacitor;

[0122] S1042. Based on the calculated DC capacitor voltage difference and output current value of the converter at the end of the current control cycle, roll over and substitute all the output voltage states of the converter in the next control cycle, and predict and calculate the capacitor output current of the converter at the end of the next control cycle.

[0123] The control objective of the grid-connected converter is to make the grid-side current ig Following the reference current i ref At steady state, the two are equal. Discretizing according to formula (1) yields:

[0124]

[0125]

[0126] From equation (7) to equation (9), we get:

[0127]

[0128] Equation (10) is the estimated expression for the capacitor current.

[0129] Let the estimated value of the grid-side current be i. gest From equation (1), its discretized value can be obtained as follows:

[0130]

[0131] The results can be calculated from equations (11) and (8).

[0132] Define the converter-side reference current as i jref From equation (3), its discretized expression is:

[0133]

[0134] The meanings of each letter in the above formula are as follows:

[0135] This is the reference current on the grid side during the current control cycle;

[0136] This is the reference current for the grid side in the previous control cycle;

[0137] This is the reference current on the grid side in the previous control cycle;

[0138] The voltage on the grid side during the current control cycle;

[0139] L g This represents the current in the filter inductor;

[0140] This represents the voltage of the filter capacitor during the current control cycle.

[0141] This is the voltage of the filter capacitor in the previous control cycle;

[0142] T s To control and sample the period;

[0143] C is the AC filter capacitor.

[0144] S1043. Based on all the output voltage states of the converter in the next control cycle, calculate the switching state corresponding to each output voltage state in the next control cycle; based on the converter output voltage collected at the previous moment and the current moment, calculate the switching state corresponding to the output voltage state at the previous moment and the current moment.

[0145] Specifically, equation (12) transforms the grid-side tracking of a given current into the converter-side current tracking of i. jref The tracking.

[0146] Converter-side output voltage u inv The expression is as follows:

[0147]

[0148] S represents the current voltage u inv It is in a switching state corresponding to different voltages.

[0149] Define the change in converter-side current as i invd Then, from equation (2), its discretized expression can be obtained as:

[0150]

[0151] Define the converter-side current estimate as i invest Then its discretized expression is:

[0152]

[0153] S1044. Based on the predicted output current of the converter and the difference between the DC capacitor current at the end of the next control cycle, the switching state corresponding to each output voltage state in the next control cycle, and the switching state corresponding to the output voltage state at the previous moment and the current moment, calculate the value of the cost function for each output voltage state in the next control cycle.

[0154] Specifically, the cost function J is defined. k Its expression is:

[0155]

[0156] In equation (16), the first term on the right corresponds to the tracking term for the given current, and the second and third terms are both terms that control the rate of change of the current, which have the function of suppressing resonance, and correspond to equation (6).

[0157] k j1 k j2 k j3These are the cost coefficients, where k j1 k j2 All are less than 1;

[0158] k j1 k j2 k j3 Adjustments are made according to specific project requirements; a larger coefficient results in stronger resonance suppression. The fourth term on the right is the switching frequency optimization term, with a cost coefficient k. j3 The larger the value, the lower the average switching frequency of the converter.

[0159] S1045. Set the converter output voltage state corresponding to the minimum cost function value as the optimal problem expression for the next switching control cycle of the converter.

[0160] Specifically, the optimal problem expression for obtaining the minimum cost function in the next switching control cycle is:

[0161]

[0162] Wherein, min(J k Minimum of the cost function, S k Indicates the on / off state of the current control cycle;

[0163] S k-1 It indicates the switching state of the previous control cycle.

[0164] S105. Based on the optimal problem expression, calculate the switching state of the converter's switching transistors in the next switching control cycle, and generate a trigger signal based on the switching state to control the opening and closing of the converter's switching devices.

[0165] The switching state of the converter in the current control cycle can be obtained from equation (17). Based on the switching state, a trigger signal is generated to control the opening and closing of the switching devices of the converter, thereby completing the control of the converter.

[0166] The control function of the grid-connected converter model predictive control method described in this application, under the premise of first order, retains the dynamic characteristics of the LCL grid-connected converter model and completes its model-based predictive control, increases the dynamic characteristics of the system, and can achieve the optimization of multiple objectives. Specifically, formulas (7) to (15) do not directly simplify the system model. Formula (16) has four cost functions, which can correspond to the control of four objectives, namely, tracking of the given current, control of the rate of change of the control current, cost coefficient, and optimization of the switching frequency.

[0167] Example 2:

[0168] This application also provides a grid-connected converter model predictive control system, which uses a grid-connected converter model predictive control method to achieve predictive control of the grid-connected converter. Please refer to... Figure 5 and Figure 6 The grid-connected converter model predictive control system includes a definition module 10, a cost function establishment module 20, an acquisition module 30, a calculation module 40, and a control output module 50 that are interconnected.

[0169] The definition module 10 is configured to define the switching control cycle of the converter;

[0170] Specifically, the definition of the converter's switching control cycle includes: defining S as the switching state, S=1, and the converter-side voltage as V. busp When S = 0, the converter-side voltage is 0; when S = -1, the converter-side voltage is -V. busn The interval between adjacent switch conduction actions is defined as the rising cycle, and the interval between adjacent switch disconnection actions is defined as the falling cycle. The arithmetic mean of the rising cycle and the falling cycle is defined as the switch control cycle.

[0171] The cost function establishment module 20 is configured to determine the voltage on the converter side according to the defined switching cycle, in order to achieve current tracking and control the rate of change of current, and to establish a cost function for controlling the switching of the switching devices of the converter.

[0172] The cost function for controlling the switching devices of the converter is as follows:

[0173]

[0174] The meanings of each letter parameter are as follows:

[0175] This is the reference current for the current control cycle on the converter side;

[0176] This is the current estimate for the current cycle on the converter side.

[0177] This is an estimated value of the current during the current control cycle on the grid side;

[0178] This represents the change in current on the converter side during the current control cycle.

[0179] k j1 k j2 k j3 These are the cost coefficients, where k j1 k j2 All are less than 1;

[0180] S k Indicates the on / off state of the current control cycle;

[0181] S k-1 It indicates the switching state of the previous control cycle.

[0182] The acquisition module 30 is configured to acquire the circuit parameter values ​​of the converter output terminal collected during the current switching control cycle.

[0183] Specifically, the circuit parameter values ​​include a grid-side reference current of i. ref The control and sampling period is T. s The voltage on the converter side is u inv The grid-side voltage is u g The grid-side current is i g The inverter-side current is i inv The capacitor current is i c .

[0184] The calculation module 40 is configured to obtain the optimal problem expression for the next switching control cycle when the cost function is minimized, based on the obtained circuit parameter values ​​of the converter output in the current switching control cycle.

[0185] The calculation module 40 includes a capacitor output current value calculation unit 401, a switch state acquisition unit 402, a cost function calculation unit 403, and an optimal calculation unit 404.

[0186] The capacitor output current value calculation unit 401 is used to calculate the output current value of the DC capacitor of the converter at the end of the current control cycle based on the circuit parameter values ​​of the converter at the current acquisition time.

[0187] Specifically, according to Figure 3 From the circuit described above, it can be seen that:

[0188]

[0189]

[0190] i c =i g -i inv (3)

[0191]

[0192] Equations (1)-(4) are the basic equations of the system shown in Figure (1). In this invention, u... g i g i inv V busp Vbusn For quantities that can be directly measured, i in the frequency domain can be obtained from formulas (1)-(4). g expression:

[0193]

[0194] Wherein, the converter-side voltage is u inv The grid-side voltage is u g The grid-side current is i g The inverter-side current is i inv The capacitor current is i c L inv Here, s represents the filter inductance on the converter side, and s denotes the Laplace operator.

[0195] To control the rate of change of the grid-side current, this invention introduces a given rate of change factor k. ref Define the current control reference value i con Its expression is:

[0196] i con =k ref i ref +(1-k ref )i g (6)

[0197] Where k ref The range is (0,1).

[0198] Discretize equation (4) to obtain the capacitor current value for the current control cycle. for:

[0199]

[0200] This is the reference value for the current in the current control cycle;

[0201] This is the capacitor current value for the current control cycle;

[0202] k ref To introduce a given rate of change factor, its range is (0,1];

[0203] i ref This is the reference current on the grid side;

[0204] i g This refers to the grid-side current.

[0205] This represents the voltage of the filter capacitor during the current control cycle.

[0206] This is the voltage of the filter capacitor in the previous control cycle;

[0207] T s To control and sample the period;

[0208] C is the AC filter capacitor;

[0209] Based on the DC capacitor voltage difference and output current value of the converter at the end of the current control cycle, the values ​​of all output voltage states of the converter in the next control cycle are rolled over and substituted to predict and calculate the capacitor output current of the converter at the end of the next control cycle.

[0210] The control objective of the grid-connected converter is to make the grid-side current i g Following the reference current i ref At steady state, the two are equal. Discretizing according to formula (1) yields:

[0211]

[0212]

[0213] From equation (7) to equation (9), we get:

[0214]

[0215] Equation (10) is the estimated expression for the capacitor current.

[0216] Let the estimated value of the grid-side current be i. gest From equation (1), its discretized value can be obtained as follows:

[0217]

[0218] The results can be calculated from equations (11) and (8).

[0219] Define the converter-side reference current as i jref From equation (3), its discretized expression is:

[0220]

[0221] This is the reference current on the grid side during the current control cycle;

[0222] This is the reference current for the grid side in the previous control cycle;

[0223] This is the reference current on the grid side in the previous control cycle;

[0224] L g This represents the current in the filter inductor;

[0225] The voltage of the filter capacitor in the current control cycle;

[0226] This is the voltage of the filter capacitor in the previous control cycle;

[0227] T s To control and sample the period;

[0228] C is the AC filter capacitor.

[0229] The switching state acquisition unit 402 is used to calculate the switching state corresponding to each output voltage state in the next control cycle based on all output voltage states of the converter in the next control cycle; and to calculate the switching state corresponding to the output voltage states in the previous and current times based on the converter output voltages collected in the previous and current times.

[0230] Specifically, equation (12) transforms the grid-side tracking of a given current into the converter-side current tracking of i. jref The tracking.

[0231] Converter-side output voltage u inv The expression is as follows:

[0232]

[0233] S represents the current voltage u inv It is in a switching state corresponding to different voltages.

[0234] Define the change in converter-side current as i invd Then, from equation (2), its discretized expression can be obtained as:

[0235]

[0236] Define the converter-side current estimate as i invest Then its discretized expression is:

[0237]

[0238] The cost function calculation unit 403 is used to calculate the value of the cost function under each output voltage state in the next control cycle based on the difference between the output current of the converter and the DC capacitor current at the end of the predicted next control cycle, the switching state corresponding to each output voltage state in the next control cycle, and the switching state corresponding to the output voltage state at the previous moment and the current moment.

[0239] Specifically, the cost function J is defined. k Its expression is:

[0240]

[0241] In equation (16), the first term on the right corresponds to the tracking term for the given current, and the second and third terms are both terms that control the rate of change of the current, which have the function of suppressing resonance, and correspond to equation (6).

[0242] k j1 k j2 k j3 These are the cost coefficients, where k j1 k j2 All are less than 1;

[0243] k j1 k j2 k j3 Adjustments are made according to specific project requirements; a larger coefficient results in stronger resonance suppression. The fourth term on the right is the switching frequency optimization term, with a cost coefficient k. j3 The larger the value, the lower the average switching frequency of the converter.

[0244] The optimal calculation unit 404 is used to calculate the optimal problem expression for the next switching control cycle of the converter based on the converter output voltage state corresponding to the minimum cost function value.

[0245] Specifically, the optimal problem expression for obtaining the minimum cost function in the next switching control cycle is:

[0246]

[0247] Wherein, min(J k Minimum of the cost function, S k Indicates the on / off state of the current control cycle;

[0248] S k-1 It indicates the switching state of the previous control cycle.

[0249] The control output module 50 is configured to calculate the switching state of the converter's switching transistors in the next switching control cycle based on the optimal output voltage, and generate a trigger signal based on the switching state to control the opening and closing of the converter's switching devices.

[0250] The switching state of the converter in the current control cycle can be obtained from equation (17). Based on the switching state, a trigger signal is generated to control the opening and closing of the switching devices of the converter, thereby completing the control of the converter.

[0251] The grid-connected converter model predictive control system described in this embodiment defines the switching control cycle of the converter through a definition module 10. A cost function establishment module 20 determines the voltage on the converter side based on the switching cycle and establishes a cost function for controlling the switching of the converter's switching devices. An acquisition module 30 acquires the circuit parameter values ​​of the converter output terminal collected during the current switching control cycle. A calculation module 40 obtains the optimal output voltage for the next switching control cycle when the cost function is minimized based on the acquired circuit parameter values ​​of the converter output terminal during the current switching control cycle. Then, a control output module 50 calculates the switching state of the converter's switching transistors during the next switching control cycle based on the optimal output voltage and generates a trigger signal to control the opening and closing of the converter's switching devices based on the switching state. The grid-connected converter model predictive control system described in this embodiment retains the dynamic characteristics of the LCL grid-connected converter model while maintaining its first-order control function and completing its model-based predictive control, thus increasing the system's dynamic characteristics and achieving optimization of multiple objectives.

[0252] The embodiments of the present invention have been described above with reference to the accompanying drawings. However, the present invention is not limited to the specific embodiments described above. The specific embodiments described above are merely illustrative and not restrictive. Those skilled in the art can make many other forms under the guidance of the present invention without departing from the spirit and scope of the claims. All of these forms are within the protection scope of the present invention.

Claims

1. A model predictive control method for grid-connected converters, characterized in that, Includes the following steps: Define the switching control cycle of the converter; The voltage on the converter side is determined according to the defined switching control cycle. Taking the LCL filter grid-connected converter as the control object, and aiming to achieve current tracking and suppress resonance by controlling the rate of change of current, a cost function for controlling the switching of the switching devices of the converter is established. The cost function includes a given current tracking term, a control current rate of change term, a cost coefficient term, and a switching frequency optimization term. The control current rate of change term has the function of suppressing resonance. The cost coefficient is adjusted according to the specific engineering requirements. The larger the coefficient, the stronger the suppression effect of resonance. The switching frequency optimization term is used to optimize the average switching frequency. The larger the cost coefficient, the lower the average switching frequency of the converter. The cost function for controlling the switching devices of the converter is as follows: ； in: This is the reference current for the current control cycle on the converter side; This is the current estimate for the current cycle on the converter side. This is an estimated value of the current during the current control cycle on the grid side; This represents the change in current on the converter side during the current control cycle. , , These are the cost coefficients, where , All are less than 1; Indicates the on / off state of the current control cycle; Indicates the switching state of the previous control cycle; Obtain the circuit parameter values ​​of the converter output terminal collected during the current switching control cycle; Based on the obtained circuit parameter values ​​at the converter output during the current switching control cycle, the optimal problem expression for the next switching control cycle when the cost function is minimized is obtained. Based on the optimal problem expression, the switching state of the converter's switching transistors in the next switching control cycle is calculated, and a trigger signal is generated based on the switching state to control the opening and closing of the converter's switching devices.

2. The model predictive control method for grid-connected converters according to claim 1, characterized in that, The defined switching control cycle of the converter includes: defining S as the switching state, S=1, and the converter-side voltage as V. busp When S=0, the converter-side voltage is 0; when S=-1, the converter-side voltage is -V. busn The interval between adjacent switch conduction actions is defined as the rising cycle, and the interval between adjacent switch disconnection actions is defined as the falling cycle. The arithmetic mean of the rising cycle and the falling cycle is defined as the switch control cycle.

3. The grid-connected converter model predictive control method as described in claim 1, characterized in that: The method for obtaining the optimal problem expression for the next switching control cycle corresponding to the minimum cost function based on the obtained circuit parameter values ​​at the converter output during the current switching control cycle includes the following steps: Based on the circuit parameter values ​​of the converter at the current acquisition time, calculate the output current value of the DC capacitor of the converter at the end of the current control cycle. Based on the DC capacitor voltage difference and output current value of the converter at the end of the current control cycle, the values ​​of all output voltage states of the converter in the next control cycle are rolled over and substituted to predict and calculate the capacitor output current of the converter at the end of the next control cycle. Based on all the output voltage states of the converter in the next control cycle, calculate the switching state corresponding to each output voltage state in the next control cycle. Based on the converter output voltage collected at the previous and current times, calculate the switching state corresponding to the output voltage state at the previous and current times; Based on the predicted output current of the converter and the difference between the DC capacitor current at the end of the next control cycle, the switching state corresponding to each output voltage state in the next control cycle, and the switching state corresponding to the output voltage state at the previous and current times, calculate the value of the cost function for each output voltage state in the next control cycle. The converter output voltage state corresponding to the minimum cost function value is set as the optimal problem expression for the next switching control cycle of the converter.

4. The model predictive control method for grid-connected converters according to claim 3, characterized in that, The optimal problem expression for obtaining the minimum cost function in the next switching control cycle is: ； in, Minimum of cost function Indicates the on / off state of the current control cycle; It indicates the switching state of the previous control cycle.

5. The model predictive control method for grid-connected converters according to claim 3, characterized in that, The calculation of the current value of the converter's output DC capacitor at the end of the current control cycle, based on the converter's circuit parameter values ​​at the current acquisition time, includes the following formula: ；（6） ；（7） This is the reference value for the current in the current control cycle; This is the capacitor current value for the current control cycle; To introduce a given rate of change factor, its range is (0,1]; This is the reference current on the grid side; This refers to the grid-side current. This represents the voltage of the filter capacitor during the current control cycle. This is the voltage of the filter capacitor in the previous control cycle; To control and sample the period; For AC filter capacitors; Based on the calculated DC capacitor voltage difference and output current value of the converter at the end of the current control cycle, the values ​​of all output voltage states of the converter in the next control cycle are substituted in a rolling manner to predict and calculate the capacitor output current of the converter at the end of the next control cycle. The discretized formula is as follows: （8） （9） The capacitor current value for the current control cycle can be obtained from formulas (7), (8), and (9). for: ；（10) This is the reference current on the grid side during the current control cycle; This is the reference current for the grid side in the previous control cycle; This is the reference current on the grid side in the previous control cycle; The voltage on the grid side during the current control cycle; This represents the current in the filter inductor; This represents the voltage of the filter capacitor during the current control cycle. This is the voltage of the filter capacitor in the previous control cycle; To control and sample the period; This is an AC filter capacitor.

6. The model predictive control method for grid-connected converters according to claim 1, characterized in that, The estimated value of the current in the current control cycle on the grid side. The expression is: （11） in, Control and adoption cycle; This is the inductance value of the filter inductor on the grid side; The voltage on the grid side during the current control cycle; The voltage of the filter capacitor in the current control cycle; This represents the current on the grid side during the previous control cycle.

7. A model predictive control system for a grid-connected converter, characterized in that, The system includes interconnected definition modules, cost function establishment modules, acquisition modules, calculation modules, and control output modules; The definition module is configured to define the switching control cycle of the converter; The cost function establishment module is configured to determine the voltage on the converter side according to the defined switching cycle, take the LCL filter grid-connected converter as the control object, and establish the cost function for controlling the switching of the switching devices of the converter with the goal of achieving current tracking and controlling the current change rate to suppress resonance balance. The cost function includes a given current tracking term, a control current rate of change term, a cost coefficient term, and a switching frequency optimization term. The control current rate of change term has the function of suppressing resonance. The cost coefficient is adjusted according to the specific engineering requirements. The larger the coefficient, the stronger the suppression effect of resonance. The switching frequency optimization term is used to optimize the average switching frequency. The larger the cost coefficient, the lower the average switching frequency of the converter. The cost function for controlling the switching devices of the converter is as follows: ； in: This is the reference current for the current control cycle on the converter side; This is the current estimate for the current cycle on the converter side. This is an estimated value of the current during the current control cycle on the grid side; This represents the change in current on the converter side during the current control cycle. , , These are the cost coefficients, where , All are less than 1; Indicates the on / off state of the current control cycle; Indicates the switching state of the previous control cycle; The acquisition module is configured to acquire circuit parameter values ​​of the converter output terminal collected during the current switching control cycle. The calculation module is configured to obtain the optimal problem expression for the next switching control cycle when the cost function is minimized, based on the obtained circuit parameter values ​​of the converter output in the current switching control cycle. The control output module is configured to calculate the switching state of the converter's switching transistors in the next switching control cycle based on the optimal problem expression, and generate a trigger signal to control the opening and closing of the converter's switching devices based on the switching state.

8. The grid-connected converter model predictive control system according to claim 7, characterized in that, The calculation module includes a capacitor output current value calculation unit, a switch state acquisition unit, a cost function calculation unit, and an optimal calculation unit. The capacitor output current calculation unit is used to calculate the output current value of the DC capacitor of the converter at the end of the current control cycle based on the circuit parameter values ​​of the converter at the current acquisition time. The switching state acquisition unit is used to calculate the switching state corresponding to each output voltage state in the next control cycle based on all output voltage states of the converter in the next control cycle. Based on the converter output voltage collected at the previous and current times, calculate the switching state corresponding to the output voltage state at the previous and current times; The cost function calculation unit is used to calculate the value of the cost function for each output voltage state in the next control cycle based on the difference between the output current of the converter and the DC capacitor current at the end of the predicted next control cycle, the switching state corresponding to each output voltage state in the next control cycle, and the switching state corresponding to the output voltage state at the previous moment and the current moment. The optimal calculation unit is used to set the converter output voltage state corresponding to the minimum cost function value as the optimal problem expression for the next switching control cycle of the converter.

Images