Model Predictive Control Method for Bipolar VSC-HVDC System Considering the Randomness of Wind Power Output

Abstract

The present invention discloses a model predictive control method for a bipolar flexible DC system considering the randomness of wind power output. Aiming at the problem that the existing wind farm - bipolar flexible DC transmission system is difficult to balance the transmitted power between the two poles in real time when the wind power output fluctuates, this method uses the model predictive control method to achieve the coordinated control of the voltage control pole and the power control pole of the bipolar flexible DC system. When the output of the wind farm fluctuates, the output power between the two poles is balanced in real time, and the system will not become unstable due to the real-time adjustment of the reference value. Compared with the existing methods, this method can realize the coordinated distribution of power between the two poles of the bipolar flexible DC transmission system while ensuring the stability of the grid voltage on the wind farm side.

Description

Technical Field

[0001] The present invention belongs to the technical field of power electronics, and particularly relates to a model predictive control method for a bipolar flexible DC system considering the randomness of wind power output. Background Art

[0002] With the gradual increase of the voltage level and transmission capacity of the flexible DC transmission system, the bipolar flexible DC transmission system has received more and more attention due to its high flexibility and reliability. The flexible DC transmission technology based on the modular multilevel converter (MMC) topology has the advantages of low manufacturing difficulty, low system loss, and high waveform quality, and has a very good application prospect in the long-distance wind power transmission.

[0003] When the bipolar MMC converter station is connected to the grid of the island wind farm, it is necessary to establish a stable frequency and voltage for the grid of the wind farm. Generally, one pole is controlled by a constant AC voltage, and the other pole is controlled by a constant power. Due to the randomness and volatility of the wind farm output, and the single-pole design capacity of the bipolar flexible DC system generally matches half of the rated capacity. If the command of the power control pole does not fully consider the uncertainty of the wind power output, the task of real-time regulating the active power output will be entirely borne by the voltage control pole. When the wind farm output fluctuates greatly, it may cause the overloading problem of the voltage control pole. For the traditional bipolar MMC control strategy based on the PI controller, if the output command of the power control pole is frequently adjusted according to the output power of the wind farm, it may lead to the instability problem of the control system; Model predictive control, as an advanced control theory, has received extensive attention in the field of power electronics in recent years. Model predictive control adjusts the MMC switching action based on the objective function, and for continuous disturbances in the system, its switching action will not switch frequently. Therefore, it is more suitable for the application scenario of the MMC control strategy considering the randomness of wind power output. At present, there is still little research on the bipolar coordinated control strategy of the sending-end MMC converter of the island wind farm - bipolar flexible DC transmission system under the fluctuation of the wind farm output. It is urgent to propose a model predictive control method for the bipolar flexible DC system considering the randomness of wind power output to ensure the safe and stable operation of the system. Summary of the Invention

[0004] The purpose of the present invention is to overcome the problem that it is difficult to balance the power transmitted between the two poles in real time when the wind power output fluctuates in the existing wind farm - bipolar flexible DC transmission system, and to provide a model predictive control method for the bipolar flexible DC system considering the randomness of wind power output, which realizes the coordinated control of the voltage control pole and the power control pole of the bipolar flexible DC system. When the wind farm output fluctuates, it can balance the output power between the two poles in real time, and because the model predictive control will not cause the instability of the system due to the real-time adjustment of the reference value, the method proposed by the present invention can ensure the stability of the system during the power regulation process.

[0005] To achieve the above-mentioned invention object, the present method adopts the following technical solutions:

[0006] A model predictive control method for a bipolar flexible DC system considering the randomness of wind power output. The control system used to implement the method includes: a positive sampling module and a negative sampling module, a positive Park transformation module and a negative Park transformation module, a positive current reference value calculation module and a negative current reference value calculation module, a positive output current prediction module and a negative output current prediction module, a positive output current objective function calculation module and a negative output current objective function calculation module, a positive internal circulating current prediction module and a negative internal circulating current prediction module, a positive internal circulating current objective function calculation module and a negative internal circulating current objective function calculation module, a positive final objective function calculation module and a negative final objective function calculation module, a positive objective function comparison and control instruction output module and a negative objective function comparison and control instruction output module;

[0007] In the positive sampling module, it includes:

[0008] A positive voltage sampling module for sampling the three-phase voltage U gabc of the MMC AC power grid;

[0009] A positive current sampling module for sampling the three-phase current I gabc1 of the positive MMC AC power grid and the internal circulating current I cabc1 of the positive MMC;

[0010] In the negative sampling module, it includes:

[0011] A negative voltage sampling module for sampling the three-phase voltage U gabc of the MMC AC power grid;

[0012] A negative current sampling module for sampling the three-phase current I gabc2 of the negative MMC AC power grid and the internal circulating current I cabc2 of the negative MMC;

[0013] The positive Park transformation module performs Park transformation on the three-phase voltage U gabc of the MMC AC power grid, the three-phase current I gabc1 of the positive AC power grid, and the internal circulating current I cabc1 of the positive to obtain the corresponding grid voltage vector U gdq in the synchronous rotating d-q coordinate system, the positive grid current vector I gdq1 and the positive internal circulating current vector I cdq1 ;

[0014] The negative Park transformation module performs Park transformation on the three-phase voltage U gabc of the MMC AC power grid, the three-phase current Igabc2 、Internal circulating current I in the negative electrode cabc2 Perform Park transformation to obtain the grid voltage vector U in the synchronous rotating d-q coordinate system gdq 、Negative electrode grid current vector I gdq2 and internal circulating current vector I in the negative electrode cdq2 ;

[0015] The reference phase θ used in the Park transformation of the positive electrode Park transformation module and the negative electrode Park transformation module r is the same;

[0016] The positive electrode current reference value calculation module controls the d-axis and q-axis components of the grid voltage vector U gdq through a PI controller to make it follow the given reference values of the d-axis and q-axis components of U gdqref U gdref and U gqref . After passing through the limiting link, the output of the PI controller is used as the reference value I of the positive electrode current gdqref1 ;

[0017] The negative electrode current reference value calculation module, based on the grid voltage vector U gdq , positive electrode grid current vector I gdq1 and negative electrode grid current vector I gdq2 , calculates the reference value I of the negative electrode current according to Equation (1) gdqref2 :

[0018]

[0019] where I gdref2 and I gqref2 are the d-axis and q-axis components of I gdqref2 respectively, U gd and U gq are the d-axis and q-axis components of U gdq respectively, I gd1 and I gq1 are the d-axis and q-axis components of I gdq1 respectively, I gd2 and I gq2 are the d-axis and q-axis components of I gdq2 respectively.

[0020] The positive electrode output current prediction module, based on the grid voltage vector U obtained in this sampling period sdq , positive electrode grid current vector I gdq1 , calculates the positive electrode grid current vector I gdq1(next) in the next sampling period when different sub-module input methods are used for the upper and lower bridge arms in this sampling period. Among them, there are N + 1 sub-module input methods in this sampling period, and I needs to be calculated N + 1 timesgdq1(next) , obtain I gdq1(next) (m), m = 1, 2, ... N + 1;

[0021] The negative - pole output current prediction module, according to the grid voltage vector U obtained in this sampling period sdq , the negative - pole grid current vector I gdq2 , respectively calculate the negative - pole grid current vector I in the next sampling period when different sub - module input methods are adopted for the upper and lower bridge arms in this sampling period gdq2(next) , where there are N + 1 sub - module input methods in this sampling period, and I needs to be calculated N + 1 times gdq2(next) , obtain I gdq2(next) (m), m = 1, 2, ... N + 1;

[0022] The positive - pole output current objective function calculation module, according to the positive - pole predicted current value I gdq1(next) (m) and the positive - pole current reference value I gdqref1 calculate the positive - pole output current objective function J 11m , m = 1, 2, ... N + 1;

[0023] The negative - pole output current objective function calculation module, according to the negative - pole predicted current value I gdq2(next) (m) and the negative - pole current reference value I gdqref2 calculate the negative - pole output current objective function J 21m , m = 1, 2, ... N + 1;

[0024] The positive - pole internal circulating current prediction module, according to the positive - pole internal circulating current I in this sampling period cdq1 , respectively calculate the positive - pole internal circulating current I in the next sampling period when N + 1 sub - module input methods are adopted for the upper and lower bridge arms in this sampling period cdq1(next) (m), m = 1, 2, ... N + 1;

[0025] The negative - pole internal circulating current prediction module, according to the negative - pole internal circulating current I in this sampling period cdq2 , respectively calculate the negative - pole internal circulating current I in the next sampling period when N + 1 sub - module input methods are adopted for the upper and lower bridge arms in this sampling period cdq2(next) (m), m = 1, 2, ... N + 1;

[0026] The positive - pole internal circulating current objective function calculation module, according to the positive - pole predicted internal circulating current I cdq1(next) (m) and the internal circulating current reference value to calculate the positive - pole internal circulating current objective function J 12m , m = 1, 2, ... N + 1, and the internal circulating current reference value is given as 0;

[0027] The negative electrode internal circulation objective function calculation module calculates the negative electrode internal circulation objective function J according to the predicted internal circulation I of the negative electrode cdq2(next) (m), where m = 1, 2,... N + 1, and the internal circulation reference value is given as 0; 22m

[0028] The positive electrode final objective function calculation module calculates the final objective function J of the positive electrode according to the positive electrode output current objective function J 11m , the positive electrode internal circulation objective function J 12m , and the weight factors p 11 , p 12 , and calculates the final objective function J of the positive electrode according to Equation (2); the negative electrode final objective function calculation module calculates the final objective function J of the negative electrode according to the negative electrode output current objective function J 1m , the negative electrode internal circulation objective function J 21m , and the weight factors p 22m , p 21 , p 22 , and calculates the final objective function J of the negative electrode according to Equation (2): 2m

[0029] J 1m = p 11 J 11m + p 12 J 12m

[0030] J 2m = p 21 J 21m + p 22 J 22m (2)

[0031] The positive electrode objective function comparison and control instruction output module compares the final objective function J of the positive electrode using the N + 1 sub-module input method, where m = 1, 2,... N + 1, and selects the sub-module input method with the smallest objective function as the control instruction for this sampling period to achieve the control of the positive electrode MMC converter; 1m

[0032] The negative electrode objective function comparison and control instruction output module compares the final objective function J of the negative electrode using the N + 1 sub-module input method, where m = 1, 2,... N + 1, and selects the sub-module input method with the smallest objective function as the control instruction for this sampling period to achieve the control of the negative electrode MMC converter. 2m

[0033] ​​​​Due to the adoption of the technical solution of the present invention, when the output of the wind farm fluctuates, the bipolar flexible DC system can ensure the stability of the grid voltage on the wind farm side while balancing the output power between the two poles in real time. Moreover, since the model predictive control will not become unstable due to the real-time adjustment of the reference value, the method proposed by the present invention can ensure the stability of the system during the power regulation process. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Figure 1 FIG. is a typical topology diagram of an island wind farm - bipolar flexible DC transmission system.

[0035] Figure 2 FIG. is a specific example structure diagram of a single - pole MMC in a bipolar flexible DC transmission system, where u ga , u gb , u gc are the grid voltages; i ga , i gb , i gc are the grid currents; u pa , u pb , u pc are the upper - arm voltages of the MMC; u na , u nb , u nc are the lower - arm voltages of the MMC; i pa , i pb , i pc are the upper - arm currents of the MMC; i na , i nb , i nc are the lower - arm currents of the MMC; U dc is the DC bus voltage, i dc is the DC bus current, L0 is the arm inductor, SM (N) is the sub - module in the MMC; N is the sub - module serial number.

[0036] Figure 3 FIG. is a schematic diagram of a specific example system of the control method of the present invention.

[0037] Figure 4 FIG. is a schematic diagram of the method for inserting N + 1 sub - modules in the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0038] In order to describe the present invention more specifically, the technical solution of the present invention will be described in detail below with reference to the drawings and specific embodiments.

[0039] The system implementation of the model predictive control method for a bipolar flexible DC system considering the randomness of wind power output in the present invention is as Figure 3As shown, it includes a positive - current reference - value calculation module 1, a negative - current reference - value calculation module 2, a positive - voltage sampling module 3, a positive - current sampling module 4, a positive - Park transformation module 5, a positive - output - current prediction module 6, a positive - output - current objective - function calculation module 7, a positive - internal - circulating - current prediction module 8, a positive - internal - circulating - current objective - function calculation module 9, a positive - final - objective - function calculation module 10, a positive - objective - function comparison and control - instruction output module 11, a negative - voltage sampling module 12, a negative - current sampling module 13, a negative - Park transformation module 14, a negative - output - current prediction module 15, a negative - output - current objective - function calculation module 16, a negative - internal - circulating - current prediction module 17, a negative - internal - circulating - current objective - function calculation module 18, a negative - final - objective - function calculation module 19, and a negative - objective - function comparison and control - instruction output module 20.

[0040] As Figure 3 shown, the model - predictive control method for a bipolar VSC - HVDC system considering the randomness of wind - power output in the present invention includes the following steps:

[0041] Collect the three - phase voltages \(U\) of the MMC AC power grid through the positive - voltage sampling module 3 and the negative - voltage sampling module 12 gabc (the same for the positive and negative poles), and respectively collect the three - phase currents \(I\) of the positive - pole MMC AC power grid through the positive - current sampling module 4 and the negative - current sampling module 13 gabc1 and the three - phase currents \(I\) of the negative - pole MMC AC power grid gabc2 , the internal circulating current \(I\) of the positive - pole MMC cabc1 and the internal circulating current \(I\) of the negative - pole MMC cabc2 .

[0042] Utilize the positive - Park transformation module 5 and the negative - Park transformation module 10 to perform Park transformation on the three - phase voltages \(U\) of the MMC AC power grid gabc , the three - phase currents \(I\) of the positive - pole AC power grid gabc1 and the three - phase currents \(I\) of the negative - pole AC power grid gabc2 , the internal circulating current \(I\) of the positive - pole cabc1 and the internal circulating current \(I\) of the negative - pole cabc2 to obtain the corresponding grid - voltage vectors \(U\) in the synchronous - rotating d - q coordinate system gdq , the positive - pole grid - current vectors \(I\) gdq1 and the negative - pole grid - current vectors \(I\) gdq2 , the positive - pole internal - circulating - current vectors \(I\) cdq1 and the negative - pole internal - circulating - current vectors \(I\) cdq2 , and the angle used for Park transformation is the reference phase \(\theta\) r .

[0043] Utilize the positive - current reference - value calculation module 1 to calculate the d - axis and q - axis components \(U\) of the grid - voltage vector \(U\) gdq and \(U\) gd and \(U\)gq It is controlled by a PI controller to follow the given reference value U gdqref of the d-axis and q-axis components U gdref and U gqref . After passing through the amplitude limiting link, the output of the PI controller is used as the reference value I of the positive electrode current gdqref1 ; among them, the d-axis voltage reference value is generally given as 1 (per unit value), the q-axis voltage reference value is generally given as 0, and the specific implementation method of the positive electrode current reference value calculation module 1 is as follows:

[0044]

[0045] where: F PId (s) and F PIq (s) are the transfer functions of the d-axis and q-axis PI controllers respectively, k pd and k pq are the proportionality coefficients of the d-axis and q-axis PI controllers respectively, k id and k iq are the integral coefficients of the d-axis and q-axis PI controllers respectively, I gdref1 , I gqref1 is the d-axis and q-axis components of I gdqref1 . The amplitude limiting value of the amplitude limiting link is set to ±I lim , and I lim is the rated operating current at the full power operating state of the system, which is 1.1 times the rated value

[0046] Using the negative electrode current reference value calculation module 2, according to the grid voltage vector U gdq , the positive electrode grid current vector I gdq1 and the negative electrode grid current vector I gdq2 , the reference value I of the negative electrode current is calculated, and the calculation method is as follows: gdqref2 The calculation method is as follows:

[0047]

[0048] where, I gdref2 and I gqref2 are the d-axis and q-axis components of I gdqref2 respectively, U gd and U gq are the d-axis and q-axis components of U gdq respectively, I gd1 and I gq1 are the d-axis and q-axis components of I gdq1 respectively, I gd2 and I gq2 are the d-axis and q-axis components of I gdq2 respectively.

[0049] Using the positive - output current prediction module 6 and the negative - output current prediction module 15, according to the grid voltage vector U obtained in this sampling period sdq , the positive - grid current vector I gdq1 and the negative - grid current vector I gdq2 , calculate respectively the positive - grid current vector I gdq1(next) and the negative - grid current vector I gdq2(next) in the next sampling period when different sub - module input methods are adopted for the upper and lower arms in this sampling period. Among them, there are N + 1 sub - module input methods in this sampling period, and it is necessary to calculate I gdq1(next) and I gdq2(next) N + 1 times to obtain I gdq1(next) (m) and I gdq2(next) (m), where m = 1, 2,... N + 1;

[0050] The N + 1 sub - module input methods are as shown in the appendix Figure 4 .

[0051] The positive - grid current vector I gdq1(next) (m) in the next sampling period is calculated according to the following method (the calculation method of the negative - grid current vector I gdq2(next) (m) is the same in principle, just replace the corresponding subscript 1 with 2):

[0052]

[0053] Among them, L is the equivalent inductance including the commutation transformer and the arm reactor, T s is the sampling period, U gd and U gq are the d - axis and q - axis components of the voltage vector U gdq respectively, I gd1 and I gq1 are the d - axis and q - axis components of I gdq1 respectively, I gd1(next) (m) and I gq1(next) (m) are the d - axis and q - axis components of I gdq1(next) (m) respectively; U nd1 (m), U nq1 (m) are the d - axis and q - axis components of the positive - lower - arm voltage when the m - th sub - module input method is adopted, and the calculation method is as follows:

[0054]

[0055] Among them, l na1 , l nb1 and l nc1 are the number of sub - modules input to the positive - corresponding - phase lower arm, and θ r is the reference phase.

[0056] U pd1 (m), U pq1 (m) are the d-axis and q-axis components of the positive upper-bridge arm voltage when the m-th sub-module input method is adopted, and their calculation methods are the same as the corresponding components of the lower-bridge arm.

[0057] Using the positive output current objective function calculation module 7 and the negative output current objective function calculation module 16, according to the positive predicted current value I gdq1(next) (m) and the negative predicted current value I gdq2(next) (m), the positive current reference value I gdqref1 and the negative current reference value I gdqref2 calculate the positive output current objective function J 11m and the negative output current objective function J 21m , m = 1, 2,... N + 1; the specific calculation method is as follows:

[0058] J 11m = |I gdref1 - I gd1(next) (m)| + |I gqref1 - I gq1(next) (m)|

[0059] J 21m = |I gdref2 - I gd2(next) (m)| + |I gqref2 - I gq2(next) (m)|

[0060] Using the positive internal circulating current prediction module 8 and the negative internal circulating current prediction module 17, according to the positive internal circulating current I cdq1 and the negative internal circulating current I cdq2 in this sampling period, calculate the positive internal circulating current I cdq1(next) (m) and the negative internal circulating current I cdq2(next) (m) in the next sampling period when the N + 1 sub-module input methods are adopted for the upper and lower bridge arms, m = 1, 2,... N + 1;

[0061] Calculate the positive internal circulating current I cdq1(next) (m) (the calculation method of the negative internal circulating current vector I cdq2(next) (m) is the same in principle, just replace the corresponding subscript 1 with 2):

[0062]

[0063] Where: T s is the sampling period, L0 is the inductance of the bridge arm reactor, I cd1 and I cq1 are Icdq1 d-axis and q-axis components of I cd1(next) (m) and I cq1(next) (m) are respectively the d-axis and q-axis components of I cdq1(next) (m), U dc1 is the DC bus voltage of the positive MMC.

[0064] Using the positive internal circulating current objective function calculation module 9 and the negative internal circulating current objective function calculation module 18, according to the predicted positive internal circulating current I cdq1(next) (m) and the predicted negative internal circulating current I cdq2(next) (m) and the internal circulating current reference value, calculate the positive internal circulating current objective function J 12m and the negative internal circulating current objective function J 22m , m = 1, 2,... N + 1; the specific calculation method is as follows:

[0065] J 12m = |I cdref1 - I cd1(next) (m)| + |I cqref1 - I cq1(next) (m)|

[0066] J 22m = |I cdref2 - I cd2(next) (m)| + |I cqref2 - I cq2(next) (m)|

[0067] Among them, I cdref1 and I cqref1 are respectively the d-axis and q-axis components of the positive internal circulating current reference value I cdqref1 , I cdref2 and I cqref2 are respectively the d-axis and q-axis components of the negative internal circulating current reference value I cdqref2 , I cdref1 , I cqref1 , I cdref2 and I cqref2 are given as 0.

[0068] Using the positive final objective function calculation module 10 and the negative final objective function calculation module 19, according to the positive output current objective function J 11m and the negative output current objective function J 21m , the positive internal circulating current objective function J 12m and the negative internal circulating current objective function J 22m , as well as the weighting factors p 11 , p 12 , p 21 , p 22 , calculate and obtain the final positive objective function J respectively1m and the final objective function J of the negative electrode 2m , and the calculation method is as follows:

[0069] J 1m = p 11 J 11m + p 12 J 12m

[0070] J 2m = p 21 J 21m + p 22 J 22m

[0071] Compare the positive electrode objective function with the control instruction output module 11 and the negative electrode objective function with the control instruction output module 20, and compare the final objective function J of the positive electrode using the N+1 sub-module input method 1m and the final objective function J of the negative electrode 2m , where m = 1, 2,... N+1, and select the sub-module input method with the minimum objective function as the control instruction for this sampling period to achieve the control of the positive electrode MMC converter and the negative electrode MMC converter.

[0072] The above description of the embodiments is to facilitate the understanding and application of the present invention by those of ordinary skill in the art. It is obvious that those skilled in the art can easily make various modifications to the above embodiments and apply the general principles described herein to other embodiments without creative labor. Therefore, the present invention is not limited to the above embodiments, and all improvements and modifications made by those skilled in the art based on the disclosure of the present invention should be within the protection scope of the present invention.

Claims

1. A model predictive control method for a bipolar flexible DC system considering the randomness of wind power output, characterized in that, The control system adopted to implement the method includes: a positive electrode sampling module and a negative electrode sampling module, a positive electrode Park transformation module and a negative electrode Park transformation module, a positive electrode current reference value calculation module and a negative electrode current reference value calculation module, a positive electrode output current prediction module and a negative electrode output current prediction module, a positive electrode output current objective function calculation module and a negative electrode output current objective function calculation module, a positive electrode internal circulating current prediction module and a negative electrode internal circulating current prediction module, a positive electrode internal circulating current objective function calculation module and a negative electrode internal circulating current objective function calculation module, a positive electrode final objective function calculation module and a negative electrode final objective function calculation module, a positive electrode objective function comparison and control instruction output module and a negative electrode objective function comparison and control instruction output module; In the positive electrode sampling module, it includes: Positive electrode voltage sampling module, sampling the three-phase voltages U of the MMC AC power grid gabc for sampling; Positive current sampling module, sampling the three-phase current I of the positive MMC AC power grid gabc1 and the internal circulating current I of the positive MMC cabc1 for sampling; In the negative electrode sampling module, it includes: Negative voltage sampling module, sampling the three-phase voltage U of the MMC AC power grid gabc for sampling; The negative - terminal current sampling module samples the three - phase currents \(I\) of the negative - terminal MMC AC power grid gabc2 and the internal circulating current \(I\) of the negative - terminal MMC cabc2 for sampling; The positive - terminal Park transformation module performs Park transformation on the three - phase voltage U of the MMC AC power grid gabc , the three - phase current I of the positive - terminal AC power grid gabc1 , the internal circulating current I of the positive terminal cabc1 to obtain the corresponding grid voltage vector U gdq , the positive - terminal grid current vector I gdq1 and the positive - terminal internal circulating current vector I cdq1 ; The negative - terminal Park transformation module performs Park transformation on the three - phase voltages U of the MMC AC power grid gabc , the three - phase currents I of the negative - terminal AC power grid gabc2 , and the internal circulating current I within the negative terminal cabc2 to obtain the corresponding grid voltage vector U in the synchronous rotating d - q coordinate system gdq , the negative - terminal grid current vector I gdq2 , and the negative - terminal internal circulating current vector I cdq2 ; The reference phase θ used for Park transformation in the positive and negative Park transformation modules is the same. r Same; The positive electrode current reference value calculation module controls the d-axis and q-axis components of the grid voltage vector U gdq through a PI controller to make them follow the given reference values U gdqref of the d-axis and q-axis components U gdref and U gqref . After passing through a limiting link, the output of the PI controller is used as the reference value I gdqref1 of the positive electrode current; The negative electrode current reference value calculation module calculates the reference value \(I_{ref}\) of the negative electrode current according to the grid voltage vector \(U\) gdq , the positive grid current vector \(I_{p}\) gdq1 and the negative grid current vector \(I_{n}\) gdq2 according to Equation (1) as follows gdqref2 : where, I gdref2 and I gqref2 are the d-axis and q-axis components of I gdqref2 respectively, U gd and U gq are the d-axis and q-axis components of U gdq respectively, I gd1 and I gq1 are the d-axis and q-axis components of I gdq1 respectively, I gd2 and I gq2 are the d-axis and q-axis components of I gdq2 respectively; The positive electrode output current prediction module calculates, according to the grid voltage vector U obtained in the current sampling period sdq , and the positive electrode grid current vector I gdq1 , the positive electrode grid current vector I gdq1(next) of the next sampling period when different sub-module input methods are adopted for the upper and lower bridge arms in the current sampling period respectively. Among them, there are N + 1 sub-module input methods in the current sampling period, and I gdq1(next) needs to be calculated N + 1 times gdq1(next) to obtain I (m), where m = 1, 2,... N + 1; The negative - pole output current prediction module calculates, based on the grid voltage vector U obtained in the current sampling period sdq , and the negative - pole grid current vector I gdq2 , the negative - pole grid current vector I gdq2(next) of the next sampling period when different sub - module input methods are adopted for the upper and lower bridge arms in the current sampling period respectively. Among them, there are N + 1 sub - module input methods in the current sampling period, and I gdq2(next) needs to be calculated N + 1 times gdq2(next) to obtain I (m), where m = 1, 2,... N + 1; The positive electrode output current objective function calculation module calculates the positive electrode output current objective function J gdq1(next) according to the positive electrode predicted current value I gdqref1 (m) and the positive electrode current reference value I 11m , where m = 1, 2,... N + 1; The negative electrode output current objective function calculation module calculates the negative electrode output current objective function J according to the negative electrode predicted current value I gdq2(next) (m) and the negative electrode current reference value I gdqref2 where m = 1, 2,... N + 1; 21m ,m = 1,2,...N+1; The positive electrode internal circulation prediction module calculates the positive electrode internal circulation I in the next sampling period when the N+1 seed modules are put into use for the upper and lower bridge arms in the current sampling period, based on the positive electrode internal circulation I in the current sampling period cdq1 respectively cdq1(next) (m), where m = 1, 2,... N+1; The negative electrode internal circulation prediction module calculates the negative electrode internal circulation I in the next sampling period when the N + 1 seed modules are put into use for the upper and lower bridge arms in the current sampling period, according to the negative electrode internal circulation I in the current sampling period cdq2 respectively cdq2(next) (m), where m = 1, 2,... N + 1; The positive electrode internal circulation objective function calculation module calculates the positive electrode internal circulation objective function J according to the predicted internal circulation I cdq1(next) (m) of the positive electrode and the internal circulation reference value, where m = 1, 2,... N + 1, and the internal circulation reference value is given as 0; 12m ​ The negative electrode internal circulation objective function calculation module calculates the negative electrode internal circulation objective function J according to the predicted internal circulation I of the negative electrode cdq2(next) (m) and the internal circulation reference value, where m = 1, 2,... N+1, and the internal circulation reference value is given as 0; 22m ​ The positive electrode final objective function calculation module calculates the final objective function \(J\) of the positive electrode according to the positive electrode output current objective function \(J\) 11m , the positive electrode internal circulation objective function \(J\) 12m , and the weight factors \(p\) 11 , \(p\) 12 , and calculates the final objective function \(J\) of the positive electrode according to Equation (2) 1m ; the negative electrode final objective function calculation module calculates the final objective function \(J\) of the negative electrode according to the negative electrode output current objective function \(J\) 21m , the negative electrode internal circulation objective function \(J\) 22m , and the weight factors \(p\) 21 , \(p\) 22 , and calculates the final objective function \(J\) of the negative electrode according to Equation (2) 2m :[[]]END J 1m = p 11 J 11m + p 12 J 12m J 2m = p 21 J 21m + p 22 J 22m (2) The positive electrode objective function comparison and control instruction output module compares the final objective function J of the positive electrode using the N+1 sub-module input method 1m , where m = 1, 2,... N+1, and selects the sub-module input method with the minimum objective function as the control instruction for this sampling period to achieve the control of the positive electrode MMC converter; The negative electrode target function comparison and control instruction output module compares the negative electrode final target function J using the N+1 sub-module input method 2m , where m = 1, 2,... N+1, selects the sub-module input method with the smallest target function as the control instruction for this sampling period, and realizes the control of the negative electrode MMC converter.

Images