Inverter network construction voltage cascade control method based on robust prediction

Abstract

The invention discloses an inverter network construction voltage cascade control method based on robust prediction, and relates to the technical field of inverter network construction voltage control. According to the method, the grid voltage concentrated disturbance and the output current concentrated disturbance of the inverter are observed in a cascading mode, an inverter output current reference and a basic voltage vector reference are established in a cascading mode, the influence of disturbance such as parameters in a grid inverter system on grid voltage control is eliminated, and the robustness and the reliability of the inverter during grid voltage control are guaranteed. According to the method, the problem caused by inconsistency of model parameters and actual parameters of the network building inverter is effectively solved, the risks of inaccurate voltage prediction and voltage waveform distortion are avoided, and meanwhile, the influence of external interference on parameter identification is reduced. In addition, the method does not need to adjust the system gain or store a large amount of data according to the change of working conditions, thereby reducing the complexity of voltage control, and improving the control performance.

Description

Technical Field

[0001] The present invention relates to the technical field of inverter network voltage control, and in particular to an inverter network voltage cascade control method based on robust prediction. Background Art

[0002] As a flexible distributed power system, the AC / DC hybrid microgrid has the ability to be flexibly connected or separated from the main power grid. The system integrates multiple components such as local renewable energy generation, inverter system, energy storage unit and load management system. In this system, the inverter uses LC filter to convert the DC power generated by renewable energy into stable AC power to achieve network construction, ensure that the energy storage and load systems can obtain stable AC power, thereby maintaining the stability of the grid voltage.

[0003] In recent years, model predictive control (MPC) has been widely used in the field of grid inverter control due to its excellent dynamic response performance and multifunctional control characteristics. MPC predicts the future state of the system based on the discrete model of the grid inverter and selects the best output state based on the cost function optimization. However, the MPC method is very sensitive to the accuracy of the filter model parameters in the grid inverter. When the model parameters are inconsistent with the actual parameters, it may cause grid voltage prediction errors, affect the inverter output performance, and even cause damage to the AC / DC hybrid microgrid.

[0004] In terms of improving the robustness of MPC parameters, scholars have developed many optimization strategies, which can be roughly summarized into two main categories: one is MPC technology based on parameter identification, and the other is data-driven MPC technology. For MPC technology based on parameter identification, researchers collect input and output data of grid-connected inverters and use specific algorithms to process these data to estimate and determine the parameters of grid-connected inverters, and then build accurate models to perform prediction and control tasks. Nevertheless, the accuracy of parameter identification is often susceptible to external interference, which may have an adverse effect on the performance of grid-connected inverters in terms of grid voltage. As for data-driven MPC technology, it collects data in real time and builds models to predict future states based on the current state and historical behavior of the system. This method does not require the establishment of an accurate mathematical model, but determines the optimal output through a value function to achieve effective control of the grid current. However, the existing data-driven MPC technology not only needs to adjust the system gain according to the changes in the working conditions of the grid-connected inverter to ensure the accuracy of the prediction, but also needs to store a large amount of current and voltage data, which may have a negative impact on the dynamic performance of the system.

[0005] Therefore, the present application aims to propose a method for cascading control of inverter grid voltage based on robust prediction to eliminate the influence of disturbances such as grid inverter parameters on grid voltage control. Summary of the invention

[0006] In view of the deficiencies in the above-mentioned background technology, the present invention proposes a cascade control method for inverter grid voltage based on robust prediction. The method observes the concentrated disturbance of inverter grid voltage and output current through cascade, establishes the inverter output current reference and basic voltage vector reference through cascade, eliminates the influence of disturbances such as parameters in the grid inverter system on the grid voltage control, and ensures the robustness and reliability of the inverter grid voltage control.

[0007] In order to achieve the above object, the technical solution adopted by the present invention is:

[0008] A method for controlling inverter grid voltage cascade based on robust prediction, the specific steps are as follows:

[0009] Step 1: According to the three-phase grid inverter topology, establish its mathematical model in the stationary coordinate system, and then establish the grid voltage prediction equation when the grid voltage concentrated disturbance exists;

[0010] Step 2: According to the grid voltage prediction equation obtained in step 1 when the grid voltage concentrated disturbance exists, a grid voltage error function is established, and the inverter output current reference value is derived;

[0011] Step 3: According to the inverter output current reference value obtained in step 2, a grid voltage concentrated disturbance observation equation is established to estimate the grid voltage concentrated disturbance in real time, thereby eliminating the influence of the concentrated disturbance on the inverter output current reference value;

[0012] Step 4: According to the three-phase grid inverter topology, establish the inverter output current prediction equation when the inverter output current concentrated disturbance exists;

[0013] Step 5: According to the inverter output current prediction equation obtained in step 4 when the inverter output current concentrated disturbance exists, establish the inverter output current error function, and derive the inverter basic voltage vector reference value;

[0014] Step 6: According to the basic voltage vector reference value of the inverter obtained in step 5, establish the concentrated disturbance observation equation of the inverter output current, estimate the concentrated disturbance of the inverter output current in real time, and then eliminate the influence of the concentrated disturbance on the basic voltage vector reference value of the inverter;

[0015] Step 7: According to the basic voltage vector reference value of the inverter obtained in step 5, a cost function is established, the optimal basic voltage vector is selected, and the three-phase modulation wave of the next control cycle is set.

[0016] This paper proposes an innovative data-driven weakly connected microgrid inverter grid voltage support control strategy. This strategy uses cascade observation technology to monitor the concentrated disturbance of the inverter grid voltage and output current, and then constructs a reference model of the inverter output current and basic voltage vector. This method effectively neutralizes the effect of parameter disturbances in the system on grid voltage control, thereby ensuring the robustness and stability of the inverter in the grid voltage control process.

[0017] Compared with the traditional model predictive control strategy, the inverter grid voltage cascade control method based on robust prediction proposed in this invention significantly improves the robustness and reliability of voltage control. It effectively solves the problem caused by the inconsistency between the model parameters and the actual parameters of the grid inverter, avoids the risk of inaccurate voltage prediction and voltage waveform distortion, and reduces the impact of external interference on parameter identification. In addition, this method does not need to adjust the system gain or store a large amount of data according to changes in working conditions, thereby reducing the complexity of voltage control and improving control performance.

[0018] The control strategy implemented in this invention successfully overcomes the limitations of traditional model predictive control in terms of parameter fluctuations, and is particularly suitable for system environments where line impedance changes frequently. This technological innovation provides a more flexible and stable effective solution for grid-forming inverters in AC / DC hybrid microgrids in the field of grid-forming voltage control. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 This is a topological diagram of the grid-connected inverter of the present invention.

[0020] Figure 2 A control block diagram for the implementation of the present invention. DETAILED DESCRIPTION

[0021] The present invention is further described in detail below in conjunction with embodiments and drawings.

[0022] Based on Figure 1 The three-phase grid-connected inverter topology structure shown in the figure, the present invention proposes a method for controlling the voltage cascade of the inverter grid based on robust prediction. Figure 2 , which mainly includes five parts: observation of grid voltage concentrated disturbance, calculation of inverter output current reference, observation of inverter output current concentrated disturbance, calculation of inverter basic voltage vector reference and evaluation of cost function. First, measure the voltage and current data of the grid inverter and observe the grid voltage concentrated disturbance; then, calculate the inverter output current reference based on the grid voltage concentrated disturbance; then, observe the inverter output current concentrated disturbance; then, calculate the inverter basic voltage vector reference based on the inverter output current concentrated disturbance; finally, evaluate the optimal basic voltage vector to be applied in the next control cycle according to the cost function. The specific detailed steps are as follows:

[0023] Step 1: According to Figure 1 The three-phase grid-connected inverter topology is shown, and its mathematical model in the stationary coordinate system is established, and then the grid-connected voltage prediction equation is established when the grid-connected voltage concentrated disturbance exists. The details are as follows:

[0024] according to Figure 1 The weakly connected microgrid inverter topology structure shown in the figure can obtain its mathematical model in the stationary coordinate system and express it as:

[0025]

[0026] Where: L m is the filter inductor model parameter, C m is the filter capacitor model parameter. i 、i l 、v i 、v c They are the inverter output current, AC load current, inverter basic voltage vector and grid voltage in the stationary coordinate system respectively.

[0027] Among them, the inverter has 8 basic voltage vectors, i∈0~7, which are v0(0,0,0), v1(1,0,0), v2(1,1,0), v3(0,1,0), v4(0,1,1), v5(0,0,1), v6(1,0,1), and v7(1,1,1).

[0028] Taking into account the load disturbance and system parameter uncertainty, the grid voltage v in formula (1) is c Further expressed as:

[0029]

[0030] Where: d vc To construct the concentrated disturbance of the grid voltage, it is expressed as:

[0031]

[0032] Where: ΔC is the parameter error between the capacitance model parameters and the actual parameters, d s is the system disturbance.

[0033] According to the front-line Euler method and the grid voltage mathematical model shown in formula (2), the grid voltage v c Discretized and expressed as:

[0034]

[0035] Where: v c [k+1] is the predicted value of the grid voltage at time k+1 in the stationary coordinate system, vc [k] is the grid voltage sampling value at time k in the stationary coordinate system, i i [k] is the inverter output current sampling value at time k in the stationary coordinate system, i l [k] is the AC load current sampling value at time k in the stationary coordinate system. vc [k] is the concentrated disturbance of the grid voltage at time k in the stationary coordinate system. s To control the cycle.

[0036] Step 2: According to the grid voltage prediction equation obtained in step 1 when there is a concentrated disturbance in the grid voltage, a grid voltage error function is established, and the reference value of the inverter output current is derived. The details are as follows:

[0037] In order to improve the robustness of grid voltage control, based on the grid voltage reference v cref and network voltage v c The error between the two designs a discrete error function, which is expressed as:

[0038] e vc [k] = ρ v [v cref [k]-v c [k]]+λ v σ v [k] (5)

[0039] Where: e vc [k] is the grid voltage error function at time k in the stationary coordinate system, ρ v is the network voltage error gain, λ v is the network voltage error integral gain. cref [k] is the grid voltage reference value at time k in the stationary coordinate system. v [k] is the integral of the grid voltage error at time k in the stationary coordinate system, which is expressed as:

[0040] σ v [k] = σ v [k-1]+[v cref [k]-v c [k]] (6)

[0041] Where: v [k-1] is the integral of the grid voltage error at time k-1 in the stationary coordinate system.

[0042] When the inverter is in steady-state operation, the grid voltage error function at time k+1 can be considered to be equal to the grid voltage error function at time k, and we get:

[0043] e vc [k+1]=e vc [k] (7)

[0044] Where: e vc [k+1] is the grid voltage error function at time k+1 in the stationary coordinate system, which is obtained by one-step forward formula (5):

[0045] e vc [k+1]=ρ v [v cref [k+1]-v c [k+1]]+λ v [σ v [k]+v cref [k+1]-v c [k+1]] (8)

[0046] Where: v cref [k+1] is the grid voltage reference value at time k+1 in the stationary coordinate system. Since the inverter generally operates at a higher power frequency, it can be considered that v cref [k+1]=v cref [k], therefore, formula (8) is further derived as:

[0047]

[0048] Substituting formula (9) and formula (5) into formula (7), we obtain:

[0049]

[0050] Where: i iref [k] is the reference value of the inverter output current at time k in the stationary coordinate system.

[0051] According to formula (10), we need to obtain the accurate network voltage concentrated disturbance d vc [k] can ensure the inverter output current reference value i iref The accuracy of [k].

[0052] Step 3: According to the inverter output current reference value obtained in step 2, establish the grid voltage concentrated disturbance observation equation, estimate the grid voltage concentrated disturbance in real time, and then eliminate the influence of the concentrated disturbance on the inverter output current reference value. The details are as follows:

[0053] In order to estimate the concentrated disturbance of grid voltage in real time vc [k], the concentrated disturbance observation equation is established according to formula (2) and expressed as:

[0054]

[0055] Where: is the estimated value of the grid voltage in the stationary coordinate system, is the estimated value of the concentrated disturbance of the grid voltage in the stationary coordinate system.

[0056] Formula (11) is discretized and expressed as:

[0057]

[0058] Where: α v is the grid voltage estimation gain, β v Estimate the gain of the concentrated disturbance of the grid voltage. is the estimated value of the grid voltage at time k+1 in the stationary coordinate system, is the estimated value of the grid voltage at time k in the stationary coordinate system. is the estimated value of the concentrated disturbance of the grid voltage at time k+1 in the stationary coordinate system, is the estimated value of the concentrated disturbance of the grid voltage at time k in the stationary coordinate system.

[0059] According to formula (12), the concentrated disturbance of grid voltage can be estimated in real time: Thus, according to formula (10), the inverter output current reference value i at time k in the stationary coordinate system is obtained: iref [k]. Since the grid-connected inverter generally works at a higher control frequency, it can be considered that i iref [k+1]=i iref [k], obtaining the inverter output current reference value effectively avoids the influence of system disturbance in the grid-connected inverter on the inverter output current reference value.

[0060] Step 4: According to Figure 1 The three-phase grid inverter topology is shown, and the inverter output current prediction equation is established when the inverter output current concentrated disturbance exists. The details are as follows:

[0061] After obtaining the reference value of the inverter output current, it is necessary to know the predicted value of the inverter output current, which is expressed as follows according to formula (1):

[0062]

[0063] Where: i i [k+1] is the predicted value of the inverter output current at time k+1 in the stationary coordinate system, i i [k] is the inverter output current sampling value at time k in the stationary coordinate system, v i [k] is the basic voltage vector of the inverter at time k in the stationary coordinate system, v c [k] is the grid voltage sampling value at time k in the stationary coordinate system.

[0064] According to formula (13), the predicted value of the inverter output current at time k+1 in the stationary coordinate system is affected by the filter inductance model parameter L mWhen the filter inductance model parameters are not equal to the actual parameters, the accuracy of the inverter output current prediction value will be affected. At this time, the inverter output current prediction value at time k+1 in the stationary coordinate system is expressed as:

[0065]

[0066] Where: d ii [k] is the concentrated disturbance of the inverter output current at time k in the stationary coordinate system.

[0067] Step 5: According to the inverter output current prediction equation obtained in step 4 when the inverter output current concentrated disturbance exists, establish the inverter output current error function, and derive the inverter basic voltage vector reference value. The details are as follows:

[0068] Based on the inverter output current reference i iref and the inverter output current v c The error between the two designs a discrete error function, which is expressed as:

[0069] e ii [k] = ρ i [i iref [k]-i i [k]]+λ i σ i [k] (15)

[0070] Where: e ii [k] is the inverter output current error function at time k in the stationary coordinate system, ρ i is the inverter output current error gain, λ v is the inverter output current error integral gain. iref [k] is the reference value of the inverter output current at time k in the stationary coordinate system. σ i [k] is the integral of the inverter output current error at time k in the stationary coordinate system, which is expressed as:

[0071] σ i [k] = σ i [k-1]+[i iref [k]-i i [k]] (16)

[0072] Where: i [k-1] is the integral of the inverter output current error at time k-1 in the stationary coordinate system.

[0073] When the inverter is in steady-state operation, it can be considered that the inverter output current error function at time k+1 is equal to the inverter output current error function at time k, and we get:

[0074] eii [k+1]=e ii [k] (17)

[0075] Where: e ii [k+1] is the inverter output current error function at time k+1 in the stationary coordinate system, which is obtained by one-step forward formula (15):

[0076] e ii [k+1]=ρ i [i iref [k+1]-i i [k+1]]+λ i [σ i [k]+i iref [k+1]-i i [k+1]] (18)

[0077] Where: i iref [k+1] is the reference value of the inverter output current at time k+1 in the stationary coordinate system. Since the inverter generally operates at a higher power frequency, it can be considered that i iref [k+1]=i iref [k], therefore, formula (18) is further derived as:

[0078]

[0079] Substituting formula (19) and formula (15) into formula (17), we obtain:

[0080]

[0081] Where: v iref [k] is the basic voltage vector reference value of the inverter at time k in the stationary coordinate system.

[0082] According to formula (20), we need to obtain the accurate inverter output current concentrated disturbance d ii [k] can ensure the basic voltage vector reference value v of the inverter iref The accuracy of [k].

[0083] Step 6: According to the basic voltage vector reference value of the inverter obtained in step 5, establish the concentrated disturbance observation equation of the inverter output current, estimate the concentrated disturbance of the inverter output current in real time, and then eliminate the influence of the concentrated disturbance on the basic voltage vector reference value of the inverter. The details are as follows:

[0084] In order to estimate the concentrated disturbance d of the inverter output current in real time ii [k], the concentrated disturbance observation equation is established according to formula (2) and expressed as:

[0085]

[0086] Where: is the estimated value of the inverter output current in the stationary coordinate system, is the estimated value of the concentrated disturbance of the inverter output current in the stationary coordinate system.

[0087] Formula (21) is discretized and expressed as:

[0088]

[0089] Where: α i is the inverter output current estimation gain, β i Estimate the gain for the inverter output current lumped disturbance. is the estimated value of the inverter output current at time k+1 in the stationary coordinate system, is the estimated value of the inverter output current at time k in the stationary coordinate system. is the estimated value of the concentrated disturbance of the inverter output current at time k+1 in the stationary coordinate system, is the estimated value of the concentrated disturbance of the inverter output current at time k in the stationary coordinate system.

[0090] According to formula (22), the concentrated disturbance of the inverter output current can be estimated in real time Thus, according to formula (20), the basic voltage vector reference value v of the inverter at time k in the stationary coordinate system is obtained: iref [k]. Since the grid-connected inverter generally works at a higher control frequency, it can be considered that v iref [k+1]=v iref [k], obtaining the basic voltage vector reference value of the inverter effectively avoids the influence of system disturbance in the grid-connected inverter on the basic voltage vector reference value of the inverter.

[0091] Step 7: According to the basic voltage vector reference value of the inverter obtained in step 5, a cost function is established to select the optimal basic voltage vector and set the three-phase modulation wave of the next control cycle. The details are as follows:

[0092] To ensure the inverter basic voltage vector prediction value v i [k+1] can track the inverter basic voltage vector reference value v iref [k+1], design cost function J i And expressed as:

[0093] J i =|v iref [k+1]-v i [k+1]| (23)

[0094] At this time, the eight basic voltage vectors of the inverter correspond to eight cost functions. The basic voltage vector with the lowest cost function is selected as the optimal vector, and the three-phase modulation wave of the next control cycle is set according to its switching state. The three-phase modulation wave can also be set in combination with space vector modulation.

[0095] The above contents are merely examples and explanations of the concept of the present invention. The technicians in this technical field may make various modifications or additions to the specific embodiments described or replace them in a similar manner. As long as they do not deviate from the concept of the invention or exceed the scope defined by the claims, they should all fall within the protection scope of the present invention.

Claims

1. A method for controlling inverter grid voltage cascade based on robust prediction, characterized in that: The specific steps are as follows: Step 1: According to the three-phase grid inverter topology, establish its mathematical model in the stationary coordinate system, and then establish the grid voltage prediction equation when the grid voltage concentrated disturbance exists; Step 2: According to the grid voltage prediction equation obtained in step 1 when the grid voltage concentrated disturbance exists, a grid voltage error function is established, and the inverter output current reference value is derived; Step 3: According to the inverter output current reference value obtained in step 2, a grid voltage concentrated disturbance observation equation is established to estimate the grid voltage concentrated disturbance in real time, thereby eliminating the influence of the concentrated disturbance on the inverter output current reference value; Step 4: According to the three-phase grid inverter topology, establish the inverter output current prediction equation when the inverter output current concentrated disturbance exists; Step 5: According to the inverter output current prediction equation obtained in step 4 when the inverter output current concentrated disturbance exists, establish the inverter output current error function, and derive the inverter basic voltage vector reference value; Step 6: According to the basic voltage vector reference value of the inverter obtained in step 5, establish the concentrated disturbance observation equation of the inverter output current, estimate the concentrated disturbance of the inverter output current in real time, and then eliminate the influence of the concentrated disturbance on the basic voltage vector reference value of the inverter; Step 7: According to the basic voltage vector reference value of the inverter obtained in step 5, a cost function is established, the optimal basic voltage vector is selected, and the three-phase modulation wave of the next control cycle is set.

2. The inverter grid voltage cascade control method based on robust prediction according to claim 1, characterized in that: Step 1 The detailed steps are as follows: According to the topological structure of the weakly connected microgrid inverter, its mathematical model in the stationary coordinate system can be obtained and expressed as: Where: L m is the filter inductor model parameter, C m is the filter capacitor model parameter; i i 、i l 、v i 、v c They are the inverter output current, AC load current, inverter basic voltage vector, and network voltage in the stationary coordinate system respectively; Among them, the inverter has 8 basic voltage vectors, i∈0~7, which are v0(0,0,0), v1(1,0,0), v2(1,1,0), v3(0,1,0), v4(0,1,1), v5(0,0,1), v6(1,0,1), v7(1,1,1); Taking into account the load disturbance and system parameter uncertainty, the grid voltage v in formula (1) is c Further expressed as: Where: d vc To construct the concentrated disturbance of the grid voltage, it is expressed as: Where: ΔC is the parameter error between the capacitance model parameters and the actual parameters, d s is the system disturbance; According to the front-line Euler method and the grid voltage mathematical model shown in formula (2), the grid voltage v c Discretized and expressed as: Where: v c [k+1] is the predicted value of the grid voltage at time k+1 in the stationary coordinate system, v c [k] is the grid voltage sampling value at time k in the stationary coordinate system, i i [k] is the inverter output current sampling value at time k in the stationary coordinate system, i l [k] is the AC load current sampling value at time k in the stationary coordinate system; d vc [k] is the concentrated disturbance of the grid voltage at time k in the stationary coordinate system; T s To control the cycle.

3. The inverter grid voltage cascade control method based on robust prediction according to claim 2, characterized in that: Step 2: The detailed steps are as follows: In order to improve the robustness of grid voltage control, based on the grid voltage reference v cref and network voltage v c The error between the two designs a discrete error function, which is expressed as: e vc [k] = ρ v [v cref [k]-v c [k]]+λ v σ v [k] (5) Where: e vc [k] is the grid voltage error function at time k in the stationary coordinate system, ρ v is the network voltage error gain, λ v is the network voltage error integral gain; v cref [k] is the reference value of the grid voltage at time k in the stationary coordinate system; σ v [k] is the integral of the grid voltage error at time k in the stationary coordinate system, which is expressed as: σ v [k]=σ v [k-1]+[vcref[k]-vc[k]](6) Where: v [k-1] is the integral of the grid voltage error at time k-1 in the stationary coordinate system; When the inverter is in steady-state operation, the grid voltage error function at time k+1 can be considered to be equal to the grid voltage error function at time k, and we get: yes vc [k+1]=e vc [k](7) Where: e vc [k+1] is the grid voltage error function at time k+1 in the stationary coordinate system, which is obtained by one-step forward formula (5): e vc [k+1]=ρ v [v cref [k+1]-v c [k+1]]+λ v [σ v [k]+v cref [k+1]-v c [k+1]] (8) Where: v cref [k+1] is the grid voltage reference value at time k+1 in the stationary coordinate system; since the inverter generally operates at a higher power frequency, it can be considered that v cref [k+1]=v cref [k], therefore, formula (8) is further derived as: Substituting formula (9) and formula (5) into formula (7), we obtain: Where: i iref [k] is the reference value of the inverter output current at time k in the stationary coordinate system; According to formula (10), we need to obtain the accurate network voltage concentrated disturbance d vc [k] can ensure the inverter output current reference value i iref The accuracy of [k].

4. The inverter grid voltage cascade control method based on robust prediction according to claim 3, characterized in that: Step 3: Detailed steps are as follows: In order to estimate the concentrated disturbance of grid voltage in real time vc [k], the concentrated disturbance observation equation is established according to formula (2) and expressed as: Where: is the estimated value of the grid voltage in the stationary coordinate system, is the estimated value of the concentrated disturbance of the grid voltage in the stationary coordinate system; Formula (11) is discretized and expressed as: Where: α v is the grid voltage estimation gain, β v To estimate the gain of concentrated disturbance of grid voltage; is the estimated value of the grid voltage at time k+1 in the stationary coordinate system, is the estimated value of the grid voltage at time k in the stationary coordinate system; is the estimated value of the concentrated disturbance of the grid voltage at time k+1 in the stationary coordinate system, is the estimated value of the concentrated disturbance of the grid voltage at time k in the stationary coordinate system; According to formula (12), the concentrated disturbance of grid voltage can be estimated in real time: Thus, according to formula (10), the inverter output current reference value i at time k in the stationary coordinate system is obtained: iref [k]; Since the grid inverter generally works at a higher control frequency, it can be considered that i iref [k+1]=i iref [k], obtaining the inverter output current reference value effectively avoids the influence of system disturbance in the grid-connected inverter on the inverter output current reference value.

5. The inverter grid voltage cascade control method based on robust prediction according to claim 4, characterized in that: Step 4: The detailed steps are as follows: After obtaining the reference value of the inverter output current, it is necessary to know the predicted value of the inverter output current, which is expressed as follows according to formula (1): Where: i i [k+1] is the predicted value of the inverter output current at time k+1 in the stationary coordinate system, i i [k] is the inverter output current sampling value at time k in the stationary coordinate system, v i [k] is the basic voltage vector of the inverter at time k in the stationary coordinate system, v c [k] is the grid voltage sampling value at time k in the stationary coordinate system; According to formula (13), the predicted value of the inverter output current at time k+1 in the stationary coordinate system is affected by the filter inductance model parameter L m When the filter inductance model parameters are not equal to the actual parameters, the accuracy of the inverter output current prediction value will be affected; at this time, the inverter output current prediction value at time k+1 in the stationary coordinate system is expressed as: Where: d ii [k] is the concentrated disturbance of the inverter output current at time k in the stationary coordinate system.

6. The inverter grid voltage cascade control method based on robust prediction according to claim 5, characterized in that: Step 5 The detailed steps are as follows: Based on the inverter output current reference i iref and the inverter output current v c The error between the two designs a discrete error function, which is expressed as: e ii [k]=ρ i [i iref [k]-i i [k]]+λ i s i [k] (15) Where: e ii [k] is the inverter output current error function at time k in the stationary coordinate system, ρ i is the inverter output current error gain, λ v is the inverter output current error integral gain; i iref [k] is the reference value of the inverter output current at time k in the stationary coordinate system; σ i [k] is the integral of the inverter output current error at time k in the stationary coordinate system, which is expressed as: s i [k]=σ i [k-1]+[i iref [k]-i i [k]] (16) Where: i [k-1] is the integral of the inverter output current error at time k-1 in the stationary coordinate system; When the inverter is in steady-state operation, it can be considered that the inverter output current error function at time k+1 is equal to the inverter output current error function at time k, and we get: yes ii [k+1]=e ii [k] (17) Where: e ii [k+1] is the inverter output current error function at time k+1 in the stationary coordinate system, which is obtained by one-step forward formula (15): e ii [k+1]=ρ i [i iref [k+1]-i i [k+1]]+λ i [s i [k]+i iref [k+1]-i i [k+1]] (18) Where: i iref [k+1] is the reference value of the inverter output current at time k+1 in the stationary coordinate system; since the inverter generally operates at a higher power frequency, it can be considered that i iref [k+1]=i iref [k], therefore, formula (18) is further derived as: Substituting formula (19) and formula (15) into formula (17), we obtain: Where: v iref [k] is the basic voltage vector reference value of the inverter at time k in the stationary coordinate system; According to formula (20), we need to obtain the accurate inverter output current concentrated disturbance d ii [k] can ensure the basic voltage vector reference value v of the inverter iref The accuracy of [k].

7. The inverter grid voltage cascade control method based on robust prediction according to claim 6, characterized in that: Step 6 The detailed steps are as follows: In order to estimate the concentrated disturbance d of the inverter output current in real time ii [k], the concentrated disturbance observation equation is established according to formula (2) and expressed as: Where: is the estimated value of the inverter output current in the stationary coordinate system, is the estimated value of the concentrated disturbance of the inverter output current in the stationary coordinate system; Formula (21) is discretized and expressed as: Where: α i is the inverter output current estimation gain, β i estimating gain for concentrated disturbance of inverter output current; is the estimated value of the inverter output current at time k+1 in the stationary coordinate system, is the estimated value of the inverter output current at time k in the stationary coordinate system; is the estimated value of the concentrated disturbance of the inverter output current at time k+1 in the stationary coordinate system, is the estimated value of the concentrated disturbance of the inverter output current at time k in the stationary coordinate system; According to formula (22), the concentrated disturbance of the inverter output current can be estimated in real time Thus, according to formula (20), the basic voltage vector reference value v of the inverter at time k in the stationary coordinate system is obtained: iref [k]; Since the grid inverter generally works at a higher control frequency, it can be considered that v iref [k+1]=v iref [k], obtaining the basic voltage vector reference value of the inverter effectively avoids the influence of system disturbance in the grid-connected inverter on the basic voltage vector reference value of the inverter.

8. The inverter grid voltage cascade control method based on robust prediction according to claim 7, characterized in that: Step 7 The detailed steps are as follows: To ensure the inverter basic voltage vector prediction value v i [k+1] can track the inverter basic voltage vector reference value v iref [k+1], design cost function J i And expressed as: J i =|v iref [k+1]-v i [k+1]| (23) At this time, the eight basic voltage vectors of the inverter correspond to eight cost functions. The basic voltage vector with the lowest cost function is selected as the optimal vector, and the three-phase modulation wave of the next control cycle is set according to its switching state. The three-phase modulation wave can also be set in combination with space vector modulation.

Images