An Optimization Method for Model Predictive Control Strategy of Three-Phase Grid-Connected Inverters

Abstract

The present invention relates to the technical field of photovoltaic grid-connected inverter, and specifically, to an optimization method for the model predictive control strategy of a three-phase grid-connected inverter. The method includes the following steps: S1. Based on the topological structure of a two-level LCL inverter, Kirchhoff's current law and Kirchhoff's voltage law, a mathematical model of the two-level LCL grid-connected inverter in the coordinate system is obtained through transformation; S2. A current prediction model is established, the differential term of the model is discretized, a cost function is constructed, and a switching times function is introduced to obtain the model predictive control strategy; S3. An event-triggered mechanism is introduced, a state space model of the inverter is established and the upper limit of the error is deduced, a trigger condition is set, and when the trigger condition is satisfied, the event mechanism is triggered to update the controller state and execute the MPC algorithm. The optimization method for the model predictive control strategy of a three-phase grid-connected inverter realizes precise control of the inductor current, ensures power quality and system stability by introducing an event-triggered mechanism and optimizing the cost function and prediction model.

Description

Technical Field

[0001] The present invention relates to the technical field of photovoltaic grid-connected inverter technology, and specifically, to an optimization method for a model predictive control strategy of a three-phase grid-connected inverter. Background Art

[0002] The existing optimization methods for the model predictive control strategy of three-phase grid-connected inverters cannot effectively balance the problems of control accuracy, switching loss, and large system operation amount. In practical applications, due to the need for frequent calculation and update of the control strategy, the system operation amount is large and the response speed is slow, thus affecting the real-time performance and stability of the overall system. Therefore, an optimization method for the model predictive control strategy of a three-phase grid-connected inverter is provided. Summary of the Invention

[0003] The purpose of the present invention is to provide an optimization method for a model predictive control strategy of a three-phase grid-connected inverter, so as to solve the problem that due to the need for frequent calculation and update of the control strategy, the system operation amount is large and the response speed is slow, thus affecting the real-time performance and stability of the overall system as proposed in the above background art.

[0004] To achieve the above purpose, the present invention aims to provide an optimization method for a model predictive control strategy of a three-phase grid-connected inverter, including the following steps:

[0005] S1. Construct a mathematical model of the grid-connected inverter. Based on the topology of the two-level LCL inverter, derive the mathematical equations of each loop through Kirchhoff's current law (KCL) and Kirchhoff's voltage law (KVL) to form a mathematical model of the grid-connected inverter, and use the transformation to obtain the mathematical model of the two-level LCL grid-connected inverter in the coordinate system.

[0006] S2. Design a model predictive control strategy. Based on the mathematical model of the grid-connected inverter, establish a current prediction model. Predict the future value of the inductor current on the inverter side by discretizing the differential terms in the mathematical model, construct a cost function in the form of squared error, and introduce a switching times function to optimize and calculate the total cost function to obtain the optimal model predictive control strategy.

[0007] S3. Introduce an event-triggered mechanism. Establish a state space model of the inverter and derive the error upper limit, set the trigger condition, and when the trigger condition is met, trigger the event mechanism, update the controller state, and execute the MPC algorithm.

[0008] As a further improvement of this technical solution, the construction of the mathematical model of the grid-connected inverter in S1 is used to describe the dynamic behavior and electrical characteristics of the inverter. The specific steps involved in constructing the mathematical model of the grid-connected inverter are as follows:

[0009] S1.1. Taking the negative potential of the DC side of the inverter as the reference potential, the output voltage u of the grid-connected inverteriN ;

[0010] S1.2. Establish the loop equations of the two-level LCL inverter according to Kirchhoff's current law (KCL) and Kirchhoff's voltage law (KVL);

[0011] S1.3. Obtain the mathematical model of the two-level LCL grid-connected inverter in the αβ coordinate system by using the Clark transformation;

[0012] Among them, the output voltage u of the grid-connected inverter in S1.1 iN The involved mathematical formulas are as follows:

[0013] u iN = S i ·U dc , i = a, b, c;

[0014]

[0015] Among them, S i is the switching function of the inverter bridge arm; U dc is the DC-side voltage;

[0016] In a three-phase balanced system, the three-phase current and voltage satisfy the following relationship:

[0017]

[0018] As a further improvement of this technical solution, the mathematical equations involved in establishing the loop equations of each loop of the two-level LCL inverter according to Kirchhoff's current law (KCL) and Kirchhoff's voltage law (KVL) in S1.2 are as follows:

[0019]

[0020] Among them, e a , e b , e c are the three-phase grid voltages; u a , u b , u c are the three-phase output voltages of the inverter; u ca , u cb , u cc are the three-phase voltages of the filter capacitor; i 1a , i 1b , i 1c are the three-phase inductor currents on the inverter side; i 2a , i 2b , i 2c are the three-phase inductor currents on the grid side; L1 and R1 are the inductor on the inverter side and the three-phase parasitic resistors respectively; L2 and R2 are the inductor on the grid side and the three-phase parasitic resistors respectively; C is the LCL filter capacitor.

[0021] As a further improvement of this technical solution, the specific mathematical model of the two-level LCL grid-connected inverter in the αβ coordinate system obtained by using the Clark transformation in S1.3 is as follows:

[0022]

[0023] Among them, e αβ is the component of the grid voltage in the αβ coordinate system; u cαβ is the component of the filter capacitor voltage in the coordinate system; i 1αβ is the component of the inductor current on the inverter side in the αβ coordinate system; i 2αβ is the component of the inductor current on the grid side in the αβ coordinate system; R2 is the parasitic resistance of the grid-side inductor; L2 is the inductance value of the grid-side inductor; is the time derivative of the grid-side inductor current in the αβ coordinate system; C is the value of the filter capacitor.

[0024] As a further improvement of this technical solution, in S2, a model predictive control strategy is designed to achieve precise control of the inverter output current and reduce the switching frequency and switching losses during the operation of the grid-connected inverter. The specific steps involved in designing the model predictive control strategy are as follows:

[0025] S2.1. Establish a current prediction model based on the mathematical model of the grid-connected inverter, and discretize the differential terms in the model formula using Euler's formula;

[0026] S2.2. Construct a cost function in the form of squared error, and establish a function f s related to the number of switchings, and construct the cost function J.

[0027] As a further improvement of this technical solution, the specific mathematical steps for discretizing i 1αβ in the model formula using Euler's formula in S2.1 are as follows:

[0028] S2.1.1. Discretize i 1αβ in the model formula using Euler's formula:

[0029]

[0030] Among them, i 1αβ (k) represents the sampled value of the inductor current i 1αβ on the inverter side at time k, and i 1αβ (k + 1) represents the value of the inductor current i 1αβ on the inverter side at time k + 1;

[0031] S2.1.2. Establish a current prediction model based on the mathematical model of the inverter in the αβ coordinate system. The discrete model of the inductor current i 1αβ on the inverter side can be obtained by using the forward difference:

[0032]

[0033] where u αβ (k) is the output voltage vector of the inverter at time k; u cαβ (k) is the voltage of the filter capacitor at time k; i 1αβ (k) is the inductor current on the inverter side; T s is the sampling period; L1 is the inductor on the inverter side;

[0034] S2.1.3. Input the filter capacitor voltage u cαβ (k), the inductor current i 1αβ (k) on the inverter side, the sampling period T s , and the inductor L1 on the inverter side, and the predicted value of the inductor current on the inverter side at time k + 1 can be obtained through calculation.

[0035] As a further improvement of this technical solution, in S2.2, a cost function in the form of a squared error is used to obtain the cost function regarding the inductor current i 1αβ (k) as follows:

[0036]

[0037] where j is the cost function in the form of the squared error of the inductor current on the inverter side; is the reference current at time k + 1 in the α coordinate system; i 1α (k + 1) is the actual current at time k + 1 in the α coordinate system; is the reference current at time k + 1 in the β coordinate system; i 1β (k + 1) is the actual current at time k + 1 in the β coordinate system;

[0038] The function f s of the number of switchings is:

[0039] f s = |S a (k) - S a (k - 1)| + |S b (k) - S b (k - 1)| + |S c (k) - S c (k - 1)|;

[0040] where f s is the function of the number of switchings; S a(k) is the switching state of phase a at time k; S a (k - 1) is the switching state of phase a at time k - 1; S b (k) is the switching state of phase b at time k; S b (k - 1) is the switching state of phase b at time k - 1; S c (k) is the switching state of phase c at time k; S c (k - 1) is the switching state of phase c at time k - 1;

[0041] The value function J is:

[0042] J = j + λ1f s ;

[0043] where J is the total value function; j is the value function in the form of the square error of the inverter - side current; f s is a function of the number of switchings; λ1 is a weight coefficient used to adjust the weight of the number of switchings in the total value function.

[0044] As a further improvement of this technical solution, an event - triggering mechanism is introduced in S3. An inverter state - space model is established and the error upper limit is derived. The triggering condition is set. When the triggering condition is met, the event - triggering mechanism is triggered, and the specific steps involved in updating the controller state and executing the MPC algorithm are as follows:

[0045] S3.1. Rearrange the mathematical model of the two - level LCL grid - connected inverter in the αβ coordinate system and establish the inverter state - space model:

[0046] x(k + 1)=Ax(k)+Bu(k);

[0047] where,

[0048]

[0049] S3.2. Select the inductor - side current i of the two - level LCL inverter 2αβ as the state variable, define the state error as ||e(t)||, and derive the upper limit of ||e(t)||;

[0050] S3.3. Introduce a flexibility coefficient ξ and determine the triggering condition.

[0051] As a further improvement of this technical solution, the mathematical calculation steps involved in defining the state error as ||e(t)|| and deriving the upper limit of ||e(t)|| in S3.2 are as follows:

[0052] S3.2.1. Define the state error as ||e(t)||:

[0053] e(t)=x(t i) -x(t) where t ∈ [t i , t i+1 );

[0054] S3.2.2. According to the inequality principle, deduce the upper limit of ||e(t)||:

[0055]

[0056] S3.2.3. From the knowledge of power electronics, it is known that the peak value E of the output phase voltage during SVPWM modulation of a three-phase inverter m should satisfy:

[0057]

[0058] where k is the modulation coefficient, 0 ≤ k ≤ 1;

[0059] S3.2.4. Combining the state-space expression of a two-level LCL grid-connected inverter and the peak value E of the output phase voltage m obtain ||u(t)|| as:

[0060]

[0061] S3.2.5. Simplify the inequality in S3.2.2:

[0062]

[0063] S3.2.6. Introduce the intermediate function θ(t):

[0064]

[0065] S3.2.7. Simplify the inequality in S3.2.5:

[0066]

[0067] S3.2.8. At time t i the specific form of θ(t) is:

[0068]

[0069] S3.2.9. Then solve the inequality in S3.2.7 and substitute the definition formula of θ(t) to obtain the ||e(t)|| inequality:

[0070]

[0071] As a further improvement of this technical solution, a flexible coefficient ξ is introduced in S3.3, and the mathematical formulas involved in determining the trigger conditions are as follows:

[0072]

[0073] When ||e(t i+1 )|| is greater than , the ET condition is activated and the MPC algorithm is executed.

[0074] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0075] 1. In this optimization method for the model predictive control strategy of a three-phase grid-connected inverter, by introducing an event-triggered mechanism, the system computational load can be effectively reduced without the need to frequently update the controller state. This mechanism triggers control updates only when the state error exceeds a preset threshold, thereby reducing the computational burden, improving the real-time performance and response speed of the control system. At the same time, it can also reduce the number of switching operations, lower the switching losses, and extend the equipment life.

[0076] 2. In this optimization method for the model predictive control strategy of a three-phase grid-connected inverter, by constructing an optimized cost function and current prediction model, accurate tracking and control of the grid-side inductor current can be achieved. The cost function in the form of squared error comprehensively considers the control accuracy and the number of switching operations, and optimizes the calculation of the total cost function, so as to select the optimal control input under various operating conditions, ensure that the grid current and voltage are of the same frequency and in phase, reduce the harmonic content, improve the grid-connected power quality, and ensure the stability and efficient operation of the system. BRIEF DESCRIPTION OF THE DRAWINGS

[0077] Figure 1 is the overall method flow chart of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0078] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0079] Embodiment:

[0080] Please refer to Figure 1 as shown, this embodiment provides an optimization method for the model predictive control strategy of a three-phase grid-connected inverter, including the following steps:

[0081] S1. Construct a mathematical model of the grid-connected inverter. Based on the topology of the two-level LCL inverter, the mathematical equations of each loop are derived through Kirchhoff's current law (KCL) and Kirchhoff's voltage law (KVL) to form a mathematical model of the grid-connected inverter, and the mathematical model of the two-level LCL grid-connected inverter in the transformed coordinate system is obtained;

[0082] In S1, a mathematical model of the grid-connected inverter is constructed to describe the dynamic behavior and electrical characteristics of the inverter. The specific steps involved in constructing the mathematical model of the grid-connected inverter are as follows:

[0083] S1.1: Taking the negative potential of the DC side of the inverter as the reference potential, the output voltage u of the grid-connected inverter iN ;

[0084] S1.2: According to Kirchhoff's current law (KCL) and Kirchhoff's voltage law (KVL), establish the loop equations of the two-level LCL inverter;

[0085] S1.3: Use the Clark transformation to obtain the mathematical model of the two-level LCL grid-connected inverter in the αβ coordinate system;

[0086] Among them, the output voltage u of the grid-connected inverter in S1.1 iN The involved mathematical formulas are as follows:

[0087] u iN = S i ·U dc , i = a, b, c;

[0088]

[0089] Among them, S i is the switching function of the inverter bridge arm; U dc is the DC side voltage;

[0090] In a three-phase balanced system, the three-phase current and voltage satisfy the following relationship:

[0091]

[0092] The mathematical equations involved in establishing the loop equations of the two-level LCL inverter according to Kirchhoff's current law (KCL) and Kirchhoff's voltage law (KVL) in S1.2 are as follows:

[0093]

[0094] Among them, e a , e b , e c are the three-phase grid voltages; u a , u b , u c are the three-phase output voltages of the inverter; u ca , u cb , u cc are the three-phase voltages of the filter capacitor; i 1a , i 1b , i 1c are the three-phase inductor currents on the inverter side; i2a 、i 2b 、i 2c is the three-phase inductor current on the grid side; L1 and R1 are the inductor on the inverter side and the three-phase parasitic resistance respectively; L2 and R2 are the grid-side inductor and the three-phase parasitic resistance respectively; C is the LCL filter capacitor.

[0095] In S1.3, the specific mathematical model of the two-level LCL grid-connected inverter in the αβ coordinate system obtained by using the Clark transformation is as follows:

[0096]

[0097] Among them, e αβ is the component of the grid voltage in the αβ coordinate system; u cαβ is the component of the filter capacitor voltage in the coordinate system; i 1αβ is the component of the inductor current on the inverter side in the αβ coordinate system; i 2αβ is the component of the inductor current on the grid side in the αβ coordinate system; R2 is the parasitic resistance of the grid-side inductor; L2 is the inductance value of the grid-side inductor; is the time derivative of the inductor current on the grid side in the αβ coordinate system; C is the filter capacitor value.

[0098] S2. Design a model predictive control strategy. Based on the mathematical model of the grid-connected inverter, establish a current prediction model. Discretize the differential terms in the mathematical model to predict the future value of the inductor current on the inverter side, construct a cost function in the form of squared error, and introduce a switching times function to optimize and calculate the total cost function to obtain the optimal model predictive control strategy;

[0099] The designed model predictive control strategy in S2 is used to achieve precise control of the inverter output current during the operation of the grid-connected inverter, reduce the switching frequency and switching losses. The specific steps involved in the designed model predictive control strategy are as follows:

[0100] S2.1. Based on the mathematical model of the grid-connected inverter, establish a current prediction model, and use Euler's formula to discretize the differential terms in the model formula;

[0101] S2.2. Construct a cost function in the form of squared error, and establish a function f s about the switching times to construct the cost function J.

[0102] In S2.1, the mathematical steps involved in discretizing i 1αβ in the model formula using Euler's formula are as follows:

[0103] S2.1.1. Discretize i 1αβ in the model formula using Euler's formula:

[0104]

[0105] Among them, i 1αβ (k) represents the sampled value of the inductor current i 1αβ on the inverter side at time k, and i 1αβ (k + 1) represents the value of the inductor current i 1αβ on the inverter side at time k + 1;

[0106] S2.1.2. Establish a current prediction model according to the mathematical model of the inverter in the αβ coordinate system. Using the forward difference, the discrete model of the inductor current i 1αβ on the inverter side can be obtained:

[0107]

[0108] Among them, u αβ (k) is the output voltage vector of the inverter at time k; u cαβ (k) is the filter capacitor voltage at time k; i 1αβ (k) is the inductor current on the inverter side; T s is the sampling period; L1 is the inductor on the inverter side;

[0109] S2.1.3. Input the filter capacitor voltage u cαβ (k), the inductor current i 1αβ (k), the sampling period T s , and the inductor L1 on the inverter side, and the predicted value of the inductor current on the inverter side at time k + 1 can be obtained through calculation.

[0110] The above S2.2 uses a value function in the form of a squared error to obtain the value function of the inductor current i 1αβ (k) as follows:

[0111]

[0112] Among them, j is the value function in the form of the squared error of the inductor current on the inverter side; is the reference current at time k + 1 in the α coordinate system; i 1α (k + 1) is the actual current at time k + 1 in the α coordinate system; is the reference current at time k + 1 in the β coordinate system; i 1β (k + 1) is the actual current at time k + 1 in the β coordinate system;

[0113] The function f s of the number of switchings is:

[0114] f s = |Sa (k)-S a (k - 1)|+|S b (k)-S b (k - 1)|+|S c (k)-S c (k - 1)|;

[0115] where f s is a function of the number of switchings; S a (k) is the switch state of phase a at time k; S a (k - 1) is the switch state of phase a at time k - 1; S b (k) is the switch state of phase b at time k; S b (k - 1) is the switch state of phase b at time k - 1; S c (k) is the switch state of phase c at time k; S c (k - 1) is the switch state of phase c at time k - 1;

[0116] The value function J is:

[0117] J = j + λ1f s ;

[0118] where J is the total value function; j is the value function in the form of the square error of the inverter - side current; f s is a function of the number of switchings; λ1 is a weight coefficient used to adjust the weight of the number of switchings in the total value function.

[0119] S3. Introduce an event - triggered mechanism, establish the inverter state - space model, deduce the error upper bound, set the trigger condition, and when the trigger condition is satisfied, trigger the event mechanism, update the controller state, and execute the MPC algorithm.

[0120] The specific steps involved in introducing the event - triggered mechanism, establishing the inverter state - space model, deducing the error upper bound, setting the trigger condition, triggering the event mechanism, updating the controller state, and executing the MPC algorithm in S3 are as follows:

[0121] S3.1. Rearrange the mathematical model of the two - level LCL grid - connected inverter in the αβ coordinate system and establish the inverter state - space model:

[0122] x(k + 1)=Ax(k)+Bu(k);

[0123] where,

[0124]

[0125] S3.2. Select the inductor - side current i of the two - level LCL inverter 2αβDefine the state variable, the state error as ||e(t)||, and derive the upper bound of ||e(t)||;

[0126] S3.3. Introduce the flexibility coefficient ξ and determine the triggering condition;

[0127] Among them, the event-triggered mechanism is a non-periodic triggering mechanism, which is suitable for control systems with limited communication bandwidth and energy. Compared with the time-triggered mechanism, the significant feature of the event-triggered mechanism is that the update of the system sampling and control algorithm does not depend on the arrival of the next cycle time, but on the deviation between the sampled value of the system state at the current moment and the preset triggering condition. If the deviation exceeds the set value, it means that the system state value at the current moment exceeds the expected value, and the system state value at this time should be updated to the controller; otherwise, the previous system state and control scheme will be maintained.

[0128] The mathematical calculation steps involved in defining the state error as ||e(t)|| and deriving the upper bound of ||e(t)|| in S3.2 are as follows:

[0129] S3.2.1. Define the state error as ||e(t)||:

[0130] e(t) = x(t i ) - x(t), t ∈ [t i , t i+1 );

[0131] S3.2.2. According to the inequality principle, derive the upper bound of ||e(t)||:

[0132]

[0133] S3.2.3. It can be known from power electronics knowledge that the peak value E of the output phase voltage during SVPWM modulation of a three-phase inverter m should satisfy:

[0134]

[0135] Among them, k is the modulation coefficient, 0 ≤ k ≤ 1;

[0136] S3.2.4. Combine the state space expression of a two-level LCL grid-connected inverter and the peak value E of the output phase voltage m to obtain ||u(t)|| as:

[0137]

[0138] S3.2.5. Simplify the inequality in S3.2.2:

[0139]

[0140] S3.2.6. Introduce the intermediate function θ(t):

[0141]

[0142] S3.2.7. Simplify the inequality in S3.2.5:

[0143]

[0144] S3.2.8. At time t i , the specific form of θ(t) is:

[0145]

[0146] S3.2.9. Then solve the inequality in S3.2.7 and substitute the definition formula of θ(t) to obtain the ||e(t)|| inequality:

[0147]

[0148] In S3.3, a flexible coefficient ξ is introduced, and the mathematical formulas involved in determining the trigger condition are as follows:

[0149]

[0150] When ||e(t i+1 )|| is greater than , the ET condition is activated and the MPC algorithm is executed;

[0151] The above shows and describes the basic principles, main features and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited by the above embodiments. The above embodiments and the descriptions in the specification are only preferred examples of the present invention and are not used to limit the present invention. Without departing from the spirit and scope of the present invention, the present invention will have various changes and improvements, and these changes and improvements all fall within the scope of the present invention claimed. The scope of protection claimed by the present invention is defined by the appended claims and their equivalents.

Claims

1. An optimization method for the model predictive control strategy of a three-phase grid-connected inverter, characterized in that: It includes the following steps: S1. Construct a mathematical model of the grid-connected inverter. Based on the topology of the two-level LCL inverter, derive the mathematical equations of each loop through Kirchhoff's Current Law (KCL) and Kirchhoff's Voltage Law (KVL) to form a mathematical model of the grid-connected inverter, and use transformation to obtain the mathematical model of the two-level LCL grid-connected inverter in the coordinate system; S2. Design a model predictive control strategy. Based on the mathematical model of the grid-connected inverter, establish a current prediction model. Discretize the differential terms in the mathematical model to predict the future value of the inductor current on the inverter side, construct a cost function in the form of squared error, and introduce a switching times function to optimize and calculate the total cost function to obtain the optimal model predictive control strategy; S3. Introduce an event-triggered mechanism. Establish a state-space model of the inverter and derive the upper limit of the error. Set the trigger condition. When the trigger condition is met, trigger the event mechanism, update the controller state, and execute the MPC algorithm; The mathematical model of the grid-connected inverter constructed in S1 is used to describe the dynamic behavior and electrical characteristics of the inverter. The specific steps involved in constructing the mathematical model of the grid-connected inverter are as follows: S1.

1. Taking the negative potential of the DC side of the inverter as the reference potential, the output voltage u of the grid-connected inverter iN ; S1.

2. According to Kirchhoff's Current Law (KCL) and Kirchhoff's Voltage Law (KVL), establish the loop equations of the two-level LCL inverter; S1.

3. Use the Clark transformation to obtain the mathematical model of the two-level LCL grid-connected inverter in the αβ coordinate system; Among them, the output voltage u of the grid-connected inverter in S1.1 iN The mathematical formulas involved are as follows: u iN = S i ·U dc , i = a, b, c; Among them, S i is the switching function of the inverter leg; U dc is the DC-side voltage; In a three-phase balanced system, the three-phase current and voltage satisfy the following relationship: The model predictive control strategy designed in S2 is used to achieve precise control of the inverter output current during the operation of the grid-connected inverter, reduce the switching frequency and switching losses. The specific steps involved in designing the model predictive control strategy are as follows: S2.

1. Based on the mathematical model of the grid-connected inverter, establish a current prediction model, and use Euler's formula to discretize the differential terms in the model formula; S2.

2. Construct a value function in the form of mean squared error and establish a function \(f\) related to the number of switchings s , and construct the value function \(J\). The value function in the form of mean squared error is adopted in S2.2 to obtain the value function about the inverter-side current i 1αβ (k) as follows: where, j is the value function in the form of the square error of the inverter-side current; is the reference current at time k+1 in the α coordinate system; i 1α (k+1) is the actual current at time k+1 in the α coordinate system; is the reference current at time k+1 in the β coordinate system; i 1β (k+1) is the actual current at time k+1 in the β coordinate system; Function f of the number of switchings s is as follows: f s = |S a (k) - S a (k - 1)| + |S b (k) - S b (k - 1)| + |S c (k) - S c (k - 1); where f s is a function of the number of switchings; S a (k) is the switching state of phase a at time k; S a (k - 1) is the switching state of phase a at time k - 1; S b (k) is the switching state of phase b at time k; S b (k - 1) is the switching state of phase b at time k - 1; S c (k) is the switching state of phase c at time k; S c (k - 1) is the switching state of phase c at time k - 1; The cost function J is: J = j + λ1f s ; Among them, J is the total value function; j is the value function in the form of the square error of the inverter-side current; f s is a function of the number of switchings; λ1 is a weight coefficient used to adjust the weight of the number of switchings in the total value function; The specific steps involved in introducing the event-triggered mechanism in S3, establishing the state-space model of the inverter, deriving the upper limit of the error, setting the trigger condition, triggering the event mechanism when the trigger condition is met, updating the controller state, and executing the MPC algorithm are as follows: S3.

1. Rearrange the mathematical model of the two-level LCL grid-connected inverter in the αβ coordinate system and establish the state-space model of the inverter: x(k + 1) = Ax(k) + Bu(k); Among them, S3.

2. Select the inductor current \(i\) of the two-level LCL inverter 2αβ as the state variable, define the state error as \(\|e(t)\|\), and derive the upper limit of \(\|e(t)\|\); S3.

3. Introduce a flexibility coefficient ξ to determine the trigger condition.

2. The method for optimizing the model predictive control strategy of a three-phase grid-connected inverter according to claim 1, characterized in that: The mathematical equations involved in establishing the loop equations of the two-level LCL inverter according to Kirchhoff's Current Law (KCL) and Kirchhoff's Voltage Law (KVL) in S1.2 are as follows: Among them, e a 、e b 、e c are the three-phase voltages of the power grid; u a 、u b 、u c are the three-phase output voltages of the inverter; u ca 、u cb 、u cc are the three-phase voltages of the filter capacitors; i 1a 、i 1b 、i 1c are the three-phase inductor currents on the inverter side; i 2a 、i 2b 、i 2c are the three-phase inductor currents on the grid side; L1 and R1 are the inductor on the inverter side and the three-phase parasitic resistors respectively; L2 and R2 are the inductor on the grid side and the three-phase parasitic resistors respectively; C is the LCL filter capacitor.

3. The optimization method for the model predictive control strategy of a three-phase grid-connected inverter according to claim 1, characterized in that: The specific mathematical model obtained by using the Clark transformation to get the mathematical model of the two-level LCL grid-connected inverter in the αβ coordinate system in S1.3 is as follows: where, e αβ is the component of the grid voltage in the αβ coordinate system; u cαβ is the component of the filter capacitor voltage in the coordinate system; i 1αβ is the component of the inductor current on the inverter side in the αβ coordinate system; i 2αβ is the component of the inductor current on the grid side in the αβ coordinate system; R2 is the parasitic resistance of the grid-side inductor; L2 is the inductance value of the grid-side inductor; is the time derivative of the grid-side inductor current in the αβ coordinate system; C is the value of the filter capacitor.

4. The optimization method for the model predictive control strategy of a three-phase grid-connected inverter according to claim 1, characterized in that: In the above S2.1, the Euler's formula is used to discretize the "i" in the model formula 1αβ The involved mathematical steps are as follows: S2.1.

1. Discretize \(i\) in the model formula using Euler's formula: 1αβ Perform discretization processing: where, i 1αβ (k) represents the sampled value of the inductor current i 1αβ on the inverter side at time k, and i 1αβ (k + 1) represents the value of the inductor current i 1αβ on the inverter side at time k + 1; S2.1.

2. Establish a current prediction model based on the mathematical model of the inverter in the αβ coordinate system. The discrete model of the inductor current i 1αβ on the inverter side can be obtained by using the forward difference: where, u αβ (k) is the output voltage vector of the inverter at time k; u cαβ (k) is the voltage of the filter capacitor at time k; i 1αβ (k) is the inductor current on the inverter side; T s is the sampling period; L1 is the inductor on the inverter side; S2.1.

3. Input the filter capacitor voltage u cαβ (k), the inductor current i 1αβ (k) on the inverter side, and the sampling period T s to the inductor L1 on the inverter side, and the predicted value of the inductor current on the inverter side at the (k + 1)th moment can be obtained through calculation.

5. The optimization method of the model predictive control strategy for a three-phase grid-connected inverter according to claim 1, characterized in that: The mathematical calculation steps involved in defining the state error as ||e(t)|| and deriving the upper limit of ||e(t)|| in S3.2 are as follows: S3.2.

1. Define the state error as ||e(t)||: e(t)=x(t i ) - x(t) t ∈ [t i , t i+1 ); S3.2.

2. According to the inequality principle, derive the upper limit of ||e(t)||: S3.2.

3. According to power electronics knowledge, the peak value E of the output phase voltage during SVPWM modulation of a three-phase inverter m should satisfy: Where k is the modulation coefficient, 0 ≤ k ≤ 1; S3.2.

4. Combine the state - space expression of the two - level LCL grid - connected inverter and the peak value E of the output phase voltage m It can be obtained that ||u(t)|| is as follows: S3.2.

5. Simplify the inequality in S3.2.2: S3.2.

6. Introduce the intermediate function θ(t): S3.2.

7. Simplify the inequality in S3.2.5: S3.2.

8. At time t i the specific form of θ(t) is as follows: S3.2.

9. Then solve the inequality in S3.2.7, and substitute the definition formula of θ(t) to obtain the ||e(t)|| inequality:

6. The method for optimizing the model predictive control strategy of a three-phase grid-connected inverter according to claim 5, characterized in that: In S3.3, a flexible coefficient ξ is introduced, and the mathematical formulas involved in determining the trigger condition are as follows: When ||e(t i+1 )|| is greater than , the ET condition is activated and the MPC algorithm is executed.

Images