An Optimized Model Predictive Control Method for a Hybrid Active Neutral-Point-Clamped Inverter

Abstract

The present invention discloses an optimized model predictive control method for a hybrid active neutral point clamped inverter. The method includes the following steps: determining the switching states of the optimized model of the hybrid active neutral point clamped inverter; determining the candidate virtual voltage vectors of the optimized model of the hybrid active neutral point clamped inverter; determining the switching sequence of the optimized model predictive control of the hybrid active neutral point clamped inverter according to the switching states and the candidate virtual voltage vectors of the optimized model of the hybrid active neutral point clamped inverter; determining the action time of the optimized model predictive control of the hybrid active neutral point clamped inverter, and combining the switching sequence of the optimized model predictive control of the hybrid active neutral point clamped inverter to perform predictive control on the optimized model. The present invention can achieve the coordinated control of the low-frequency unit and the high-frequency unit in the optimized model and the capacitor voltage balance at a fixed switching frequency, and at the same time shorten the operation time of the model predictive control method.

Description

Technical Field

[0001] The present invention relates to the field of model predictive control, and particularly to an optimized model predictive control method for a hybrid active neutral point clamped inverter. Background Art

[0002] The optimization model of the hybrid active neutral point clamped inverter is specifically a 4SiC-2Si hybrid three-level active neutral point clamped inverter. Each phase leg of this inverter is composed of 4 SiC MOSFETs and 2 Si IGBTs. Only the SiC MOSFETs operate at a high switching frequency. Compared with the all-SiC three-level active neutral point clamped inverter, this inverter also achieves high efficiency and low cost. At the same time, the balance degree of device loss distribution of the 4-SiC three-level active neutral point clamped inverter is improved. The key to the efficient operation of the Si / SiC hybrid topology is to control the high-frequency operation of the SiC MOSFETs and the low-frequency operation of the Si IGBTs to achieve frequency decoupling control of different power devices.

[0003] Model predictive control methods have received extensive attention in the fields of power electronics and AC drives because they can easily consider non-linear multivariable constraints. Traditional model predictive control methods consider the switching states of the converter as a whole, making it difficult to achieve decoupled operation of the switching devices in the Si / SiC hybrid topology. At the same time, the changing switching frequency, heavy computational burden, and difficult-to-tune evaluation function weight factors have hindered the widespread application of traditional model predictive control methods.

[0004] Currently, existing model predictive control methods decompose the five-level active neutral point clamped inverter into a low-frequency unit and a high-frequency unit, and propose a hybrid model predictive control method to ensure the fundamental frequency operation of the low-frequency unit and achieve fixed switching frequency operation of the high-frequency unit. In the hybrid model predictive control, classical model predictive control adjusts the low-frequency unit, while the duty cycle optimization model predictive control adjusts the high-frequency unit. At the same time, capacitor voltage balance is achieved by adjusting the duty cycle, thus avoiding the process of adjusting the weight factor. Between two voltage sampling operations, the change value of the capacitor voltage is estimated by measuring the current flowing through the sub-module capacitor. However, the existing model predictive control methods are only applicable to the five-level active neutral point clamped inverter. The balance of the capacitor voltage is achieved by adjusting the switching duty cycle, which will increase the THD of the system output current and reduce the control accuracy and rapidity. Summary of the Invention

[0005] Aiming at the above deficiencies in the prior art, the present invention provides an optimized model predictive control method for a hybrid active neutral point clamped inverter, which can achieve the coordinated control of the low-frequency unit and the high-frequency unit in the optimization model and the capacitor voltage balance at a fixed switching frequency, and at the same time shorten the operation time of the model predictive control method.

[0006] To achieve the above-mentioned invention purpose, the technical solution adopted by the present invention is as follows:

[0007] An optimized model predictive control method for a hybrid active neutral point clamped inverter, comprising the following steps:

[0008] S1. Determine the switching states of the optimized model of the hybrid active neutral point clamped inverter according to the topological structure of the optimized model of the hybrid active neutral point clamped inverter;

[0009] S2. Calculate the evaluation function of the optimized model of the hybrid active neutral point clamped inverter by using the forward Euler method, and determine the candidate virtual voltage vectors of the optimized model of the hybrid active neutral point clamped inverter according to the evaluation function of the optimized model of the hybrid active neutral point clamped inverter;

[0010] S3. Determine the switching sequence of the optimized model predictive control of the hybrid active neutral point clamped inverter according to the switching states of the optimized model of the hybrid active neutral point clamped inverter in step S1 and the candidate virtual voltage vectors of the optimized model of the hybrid active neutral point clamped inverter in step S2;

[0011] S4. Determine the action time of the optimized model predictive control of the hybrid active neutral point clamped inverter according to the reference voltage of the optimized model of the hybrid active neutral point clamped inverter, and perform predictive control on the optimized model of the hybrid active neutral point clamped inverter in combination with the switching sequence of the optimized model predictive control of the hybrid active neutral point clamped inverter in step S3.

[0012] Further, step S2 includes the following steps:

[0013] S21. Calculate the output voltage of the optimized model of the hybrid active neutral point clamped inverter in the two-phase stationary coordinate system;

[0014] S22. Calculate the predicted value of the output voltage of the optimized model of the hybrid active neutral point clamped inverter by using the forward Euler method according to the output voltage of the optimized model of the hybrid active neutral point clamped inverter in the two-phase stationary coordinate system in step S21;

[0015] S23. Calculate the evaluation function of the optimized model of the hybrid active neutral point clamped inverter according to the predicted value of the output voltage of the optimized model of the hybrid active neutral point clamped inverter in step S22;

[0016] S24. Determine the candidate virtual voltage vectors of the optimized model of the hybrid active neutral point clamped inverter according to the evaluation function of the optimized model of the hybrid active neutral point clamped inverter in step S23.

[0017] Further, in step S22, the predicted output voltage of the optimization model of the hybrid active neutral point clamped inverter is calculated using the forward Euler method, expressed as:

[0018]

[0019] Where: U αβ (k + 1) is the predicted output voltage of the optimization model of the hybrid active neutral point clamped inverter at the (k + 1)-th moment, L is the load inductor of the optimization model of the hybrid active neutral point clamped inverter, is the reference current of the optimization model of the hybrid active neutral point clamped inverter, i αβ (k) is the current sampling value of the optimization model of the hybrid active neutral point clamped inverter at the k-th moment, T s is the sampling period, and R is the load resistance of the optimization model of the hybrid active neutral point clamped inverter.

[0020] Further, in step S23, the evaluation function of the optimization model of the hybrid active neutral point clamped inverter is calculated, expressed as:

[0021]

[0022] Where: g V is the evaluation function value of the optimization model of the hybrid active neutral point clamped inverter, is the abscissa of the output space voltage vector of the optimization model of the hybrid active neutral point clamped inverter, u α (k + 1) is the abscissa of the predicted output voltage of the optimization model of the hybrid active neutral point clamped inverter at the (k + 1)-th moment, is the ordinate of the output space voltage vector of the optimization model of the hybrid active neutral point clamped inverter, u β (k + 1) is the ordinate of the predicted output voltage of the optimization model of the hybrid active neutral point clamped inverter at the (k + 1)-th moment.

[0023] Further, step S3 includes the following sub-steps:

[0024] S31. According to the switching state of the optimization model of the hybrid active neutral point clamped inverter in step S1, determine the switching sequence that meets the coordinated control of the low-frequency unit and the high-frequency unit in the optimization model of the hybrid active neutral point clamped inverter;

[0025] S32. According to the candidate virtual voltage vectors of the optimization model of the hybrid active neutral point clamped inverter in step S2, determine the switching sequence that meets the capacitor voltage balance in the optimization model of the hybrid active neutral point clamped inverter;

[0026] S33. Calculate the output voltage prediction difference of the optimization model of the hybrid active neutral point clamped inverter;

[0027] S34. Determine the switching sequence of the predictive control of the optimization model of the hybrid active neutral point clamped inverter according to the switching sequence that meets the coordinated control of the low-frequency unit and the high-frequency unit in the optimization model of the hybrid active neutral point clamped inverter in step S31, the switching sequence that meets the capacitor voltage balance in the optimization model of the hybrid active neutral point clamped inverter in step S32, and the output voltage prediction difference of the optimization model of the hybrid active neutral point clamped inverter in step S33.

[0028] Further, in step S33, the calculation of the output voltage prediction difference of the optimization model of the hybrid active neutral point clamped inverter is expressed as:

[0029]

[0030] where: ΔU c (k + 1) is the difference in the DC-side capacitor voltage of the optimization model of the hybrid active neutral point clamped inverter at the (k + 1)-th moment, C is the value of the DC-side capacitor of the optimization model of the hybrid active neutral point clamped inverter, i is a constant, i = (1, 2, 3), t i is the action time of different switching vectors in the switching sequence, S = [S a S b S c , S a is the switching state of phase a, S b is the switching state of phase b, S c is the switching state of phase c, T is the transpose symbol, i abc = [i a i b i c , i a is the current of phase a, i b is the current of phase b, i c is the current of phase c, ΔU c (k) is the difference in the DC-side capacitor voltage of the optimization model of the hybrid active neutral point clamped inverter at the k-th moment.

[0031] Further, in step S4, according to the reference voltage of the optimization model of the hybrid active neutral point clamped inverter, determine the action time of the predictive control of the optimization model of the hybrid active neutral point clamped inverter, which is expressed as:

[0032]

[0033] where: t 1 is the action time of the voltage vector v a , v ais the first voltage vector in the switching sequence, t 2 is the action time of the voltage vector v b , v b is the second voltage vector in the switching sequence, t 3 is the action time of the voltage vector v c , v c is the third voltage vector in the switching sequence, v * is the reference voltage of the optimization model of the hybrid active neutral-point clamped inverter, T s is the sampling period.

[0034] The beneficial effects of the present invention are as follows:

[0035] (1) The present invention uses the forward Euler method to calculate the predicted output voltage value of the optimization model of the hybrid active neutral-point clamped inverter, and then calculates the evaluation function of the optimization model of the hybrid active neutral-point clamped inverter. Then, the candidate virtual voltage vectors of the optimization model of the hybrid active neutral-point clamped inverter are determined, and combined with the switching state of the optimization model of the hybrid active neutral-point clamped inverter, the switching sequence of the predictive control of the optimization model of the hybrid active neutral-point clamped inverter is determined, which can realize the coordinated control of the low-frequency unit and the high-frequency unit in the optimization model and the capacitor voltage balance at a fixed switching frequency;

[0036] (2) The present invention determines the action time of the predictive control of the optimization model of the hybrid active neutral-point clamped inverter through the reference voltage of the optimization model of the hybrid active neutral-point clamped inverter, and then combines the switching sequence of the predictive control of the optimization model of the hybrid active neutral-point clamped inverter to perform predictive control on the optimization model of the hybrid active neutral-point clamped inverter, which can realize the coordinated control of the low-frequency unit and the high-frequency unit in the optimization model and the capacitor voltage balance at a fixed switching frequency, and at the same time shorten the running time of the model predictive control method. Description of the Drawings

[0037] Figure 1 is a schematic flow chart of a predictive control method for an optimization model of a hybrid active neutral-point clamped inverter;

[0038] Figure 2 is a schematic topological structure diagram of an optimization model of a hybrid active neutral-point clamped inverter;

[0039] Figure 3 is a schematic derivation diagram of an optimization model of a hybrid active neutral-point clamped inverter;

[0040] Figure 4 is a schematic diagram of the distribution of candidate virtual voltage vectors and sector division of an optimization model of a hybrid active neutral-point clamped inverter;

[0041] Figure 5 is the sub-sector sI-3 Effect of medium voltage large vector PNN on capacitor voltage;

[0042] Figure 6 For sub-sector s I-3 Effect of medium voltage medium vector PON on capacitor voltage;

[0043] Figure 7 For sub-sector s I-3 Effect of medium voltage small P-type vector PPO on capacitor voltage;

[0044] Figure 8 For sub-sector s I-3 Effect of medium voltage small N-type vector PON on capacitor voltage. Specific embodiments

[0045] The following describes the specific embodiments of the present invention to facilitate those skilled in the art of the present technology to understand the present invention. However, it should be clear that the present invention is not limited to the scope of the specific embodiments. For those of ordinary skill in the art of the present technology, as long as various changes are within the spirit and scope of the present invention defined and determined by the appended claims, these changes are obvious, and all inventions and creations using the concept of the present invention are within the scope of protection.

[0046] As Figure 1 shown, an optimized model predictive control method for a hybrid active neutral point clamped inverter includes steps S1 - S4, specifically as follows:

[0047] S1. Determine the switching states of the optimized model of the hybrid active neutral point clamped inverter according to the topological structure of the optimized model of the hybrid active neutral point clamped inverter.

[0048] In an alternative embodiment of the present invention, the present invention determines the switching states of the optimized model of the hybrid active neutral point clamped inverter according to the topological structure of the optimized model of the hybrid active neutral point clamped inverter.

[0049] As Figure 2 shown, the present invention provides a schematic diagram of the topological structure of the optimized model of the hybrid active neutral point clamped inverter. As Figure 3 shown, the 3L - ANPC inverter topological structure is formed by different switching tubes being turned on and off, connecting the AC side output terminal to different DC side voltage levels to form an output current path. Therefore, the two decoupled parts of the circuit and different types of devices can be used to form a high-frequency unit and a low-frequency unit. One is the selection of high-frequency voltage levels, and the other is the low-frequency phase change. The level selection switch unit can achieve different voltage levels (V dc / 2, 0, -V dcThe commutation switch unit, which switches between (N, P) and (P, N), is similar to a single-pole double-throw switch and, in cooperation with the high-frequency level selection switch unit, realizes the voltage and current output between the N terminal and the P terminal. In this way, the twelve switches of the 3L-ANPC converter are replaced by SiC MOSFETs, and the other switches still use SiIGBTs.

[0050] The present invention determines the switching states of the optimization model of the hybrid active neutral-point clamped inverter as follows:

[0051] Table I Switching States of 3L-ANPC Inverter

[0052]

[0053] Table I lists the switching states of each phase. Since there are redundancies in the optimization model of the hybrid active neutral-point clamped inverter, there are four zero-switching states: OL1, OL2, OU1, and OU2. The negative switching state and the positive switching state are respectively referred to as N and P. The modulation strategy in which only two zero-switching vectors (i.e., OU2 and OL2) are active is more conducive to extending the life of the device. The model predictive control method in the present invention only uses two zero-switching states (hereinafter, OU2 is denoted as U and OL2 is denoted as L).

[0054] S2. Calculate the evaluation function of the optimization model of the hybrid active neutral-point clamped inverter by the forward Euler method, and determine the candidate virtual voltage vectors of the optimization model of the hybrid active neutral-point clamped inverter according to the evaluation function of the optimization model of the hybrid active neutral-point clamped inverter.

[0055] In an alternative embodiment of the present invention, the present invention calculates the evaluation function of the optimization model of the hybrid active neutral-point clamped inverter by the forward Euler method, and determines the candidate virtual voltage vectors of the optimization model of the hybrid active neutral-point clamped inverter according to the evaluation function of the optimization model of the hybrid active neutral-point clamped inverter.

[0056] Step S2 includes the following steps:

[0057] S21. Calculate the output voltage of the optimization model of the hybrid active neutral-point clamped inverter in the two-phase stationary coordinate system.

[0058] The present invention calculates the output voltage of the optimization model of the hybrid active neutral-point clamped inverter in the two-phase stationary coordinate system, expressed as:

[0059]

[0060] Where: U αβ is the output voltage of the optimization model of the hybrid active neutral-point clamped inverter in the two-phase stationary coordinate system, L is the load inductor of the optimization model of the hybrid active neutral-point clamped inverter, iαβ is the current sampling value of the optimized model of the hybrid active neutral point clamped inverter, t is the sampling time, and R is the load resistance of the optimized model of the hybrid active neutral point clamped inverter.

[0061] S22. According to the output dynamic voltage of the optimized model of the hybrid active neutral point clamped inverter in the two-phase stationary coordinate system in step S21, use the forward Euler method to calculate the predicted output voltage value of the optimized model of the hybrid active neutral point clamped inverter.

[0062] The present invention uses the forward Euler method to calculate the predicted output voltage value of the optimized model of the hybrid active neutral point clamped inverter, expressed as:

[0063]

[0064] where: U αβ (k + 1) is the predicted output voltage value of the optimized model of the hybrid active neutral point clamped inverter at the (k + 1)-th moment, L is the load inductance of the optimized model of the hybrid active neutral point clamped inverter, is the reference current of the optimized model of the hybrid active neutral point clamped inverter, i αβ (k) is the current sampling value of the optimized model of the hybrid active neutral point clamped inverter at the k-th moment, T s is the sampling period, and R is the load resistance of the optimized model of the hybrid active neutral point clamped inverter.

[0065] S23. According to the predicted output voltage value of the optimized model of the hybrid active neutral point clamped inverter in step S22, calculate the evaluation function of the optimized model of the hybrid active neutral point clamped inverter.

[0066] The present invention calculates the evaluation function of the optimized model of the hybrid active neutral point clamped inverter, expressed as:

[0067]

[0068] where: g V is the evaluation function value of the optimized model of the hybrid active neutral point clamped inverter, is the abscissa of the output space voltage vector of the optimized model of the hybrid active neutral point clamped inverter, u α (k + 1) is the abscissa of the predicted output voltage value of the optimized model of the hybrid active neutral point clamped inverter at the (k + 1)-th moment, is the ordinate of the output space voltage vector of the optimized model of the hybrid active neutral point clamped inverter, u β (k + 1) is the ordinate of the predicted output voltage value of the optimized model of the hybrid active neutral point clamped inverter at the (k + 1)-th moment.

[0069] S24. Determine the candidate virtual voltage vectors of the optimization model of the hybrid active neutral point clamped inverter according to the evaluation function of the optimization model of the hybrid active neutral point clamped inverter in step S23.

[0070] In order to reduce the computational burden, according to the evaluation function of the optimization model of the hybrid active neutral point clamped inverter, a two-stage process method is adopted to determine the candidate virtual voltage vectors of the optimization model of the hybrid active neutral point clamped inverter. As Figure 4 shown, the first stage is to select one from the six medium voltage vectors (v 7 , v 8 , v 9 , v 10 , v 11 , v 12 ) to minimize the evaluation function of the optimization model of the hybrid active neutral point clamped inverter to determine the optimal voltage vector. After determining the optimal voltage vector, the virtual voltage vector will be in the same sector as the optimal voltage vector. As Figure 4 shown, the geometric midpoint of the triangle is the candidate virtual voltage vector in different sectors.

[0071] S3. Determine the switching sequence of the predictive control of the optimization model of the hybrid active neutral point clamped inverter according to the switching state of the optimization model of the hybrid active neutral point clamped inverter in step S1 and the candidate virtual voltage vectors of the optimization model of the hybrid active neutral point clamped inverter in step S2.

[0072] In an optional embodiment of the present invention, the present invention determines the switching sequence of the predictive control of the optimization model of the hybrid active neutral point clamped inverter according to the switching state of the optimization model of the hybrid active neutral point clamped inverter and the candidate virtual voltage vectors of the optimization model of the hybrid active neutral point clamped inverter.

[0073] Step S3 includes the following sub-steps:

[0074] S31. Determine the switching sequence that meets the coordinated control of the low-frequency unit and the high-frequency unit in the optimization model of the hybrid active neutral point clamped inverter according to the switching state of the optimization model of the hybrid active neutral point clamped inverter in step S1.

[0075] The optimization model of the hybrid active neutral point clamped inverter has more redundant vectors. Using inappropriate vectors or sequences will increase the losses of Si IGBTs, resulting in an increase in the device junction temperature, a decrease in the converter efficiency, and ultimately affecting the performance of the converter. Therefore, in the present invention, the influence of switching losses is considered when selecting redundant switching states. The switching losses of Si IGBTs are the lowest (constituting the low-frequency unit), while SiC MOSFETs undertake all the switching (constituting the high-frequency unit). Therefore, according to the switching states of the optimization model of the hybrid active neutral point clamped inverter, the present invention can determine that the switches only operate in the P state and the U state, or only in the N state and the L state within one control period. Therefore, during the half-wave period of the fundamental voltage (positive half-cycle or negative half-cycle), only S x1 、S x2 、S x3 and S x4 in the SiC MOSFETs can act quickly. The Si IGBTs only function during voltage commutation (the switching action is between the U and L states). Therefore, the present invention obtains the five-segment switching sequence of the first sector as follows:

[0076] Table II Five-segment switching sequence of the first sector

[0077]

[0078] S32. According to the candidate virtual voltage vectors of the optimization model of the hybrid active neutral point clamped inverter in step S2, determine the switching sequence that meets the capacitor voltage balance in the optimization model of the hybrid active neutral point clamped inverter.

[0079] In view of the fact that the redundant zero voltage vectors of the active neutral point clamped inverter have no influence on the DC-side capacitor voltage, the neutral point clamped converter provides an example for an effective method of using different voltage vectors to control the capacitor voltage. In the traditional model control method, taking the sub-sector s Figure 4 in I-3 as an example. The large vector PNN does not affect the NP voltage, as shown in Figure 5 . The medium vector PON will cause capacitor voltage imbalance, as shown in Figure 6 . The small P-type vector PPO discharges the upper capacitor C 1 and charges the lower capacitor C 2 , as shown in Figure 7 . On the contrary, the small N-type vector PON can discharge the lower capacitor C2 and charge the upper capacitor C 1 , as shown in Figure 8As shown in the figure. Based on this, the present invention can divide the sub-sectors into two different switching sequences, namely N-type and P-type, according to the different effects of the vectors in the sub-sectors on the capacitor voltage, and can select the sequence according to the prediction of the capacitor voltage in the next control period. Therefore, the present invention determines the switching sequence that meets the capacitor voltage balance in the optimization model of the hybrid active neutral point clamped inverter according to the candidate virtual voltage vectors of the optimization model of the hybrid active neutral point clamped inverter.

[0080] S33. Calculate the predicted output voltage difference of the optimization model of the hybrid active neutral point clamped inverter.

[0081] The present invention calculates the predicted output voltage difference of the optimization model of the hybrid active neutral point clamped inverter, expressed as:

[0082]

[0083] Where: ΔU c (k + 1) is the difference in the DC-side capacitor voltage of the optimization model of the hybrid active neutral point clamped inverter at the (k + 1)-th moment, C is the DC-side capacitor value of the optimization model of the hybrid active neutral point clamped inverter, i is a constant, i = (1, 2, 3), t i is the action time of different switching vectors in the switching sequence, S = [S a S b S c , S a is the switching state of phase a, S b is the switching state of phase b, S c is the switching state of phase c, T is the transpose symbol, i abc = [i a i b i c , i a is the current of phase a, i b is the current of phase b, i c is the current of phase c, ΔU c (k) is the difference in the DC-side capacitor voltage of the optimization model of the hybrid active neutral point clamped inverter at the k-th moment.

[0084] S34. According to the switching sequence that meets the coordinated control of the low-frequency unit and the high-frequency unit in the optimization model of the hybrid active neutral point clamped inverter in step S31, the switching sequence that meets the capacitor voltage balance in the optimization model of the hybrid active neutral point clamped inverter in step S32, and the predicted output voltage difference of the optimization model of the hybrid active neutral point clamped inverter in step S33, determine the switching sequence of the predicted control of the optimization model of the hybrid active neutral point clamped inverter.

[0085] S4. Determine the action time of the model predictive control of the hybrid active neutral point clamped inverter according to the reference voltage of the optimization model of the hybrid active neutral point clamped inverter, and perform model predictive control on the optimization model of the hybrid active neutral point clamped inverter in combination with the switching sequence of the model predictive control of the hybrid active neutral point clamped inverter in step S3.

[0086] In an alternative embodiment of the present invention, the present invention determines the action time of the model predictive control of the hybrid active neutral point clamped inverter according to the reference voltage of the optimization model of the hybrid active neutral point clamped inverter, and combines the switching sequence of the model predictive control of the hybrid active neutral point clamped inverter to perform model predictive control on the optimization model of the hybrid active neutral point clamped inverter.

[0087] In each sub-sector of the present invention, the action time of each voltage space vector is calculated by synthesizing the reference voltage v using the three nearest static voltage vectors and volt-second balance. * .

[0088] The present invention determines the action time of the model predictive control of the hybrid active neutral point clamped inverter according to the reference voltage of the optimization model of the hybrid active neutral point clamped inverter, which is expressed as:

[0089]

[0090] where: t 1 is the action time of the voltage vector v a , v a is the first voltage vector in the switching sequence, t 2 is the action time of the voltage vector v b , v b is the second voltage vector in the switching sequence, t 3 is the action time of the voltage vector v c , v c is the third voltage vector in the switching sequence, v * is the reference voltage of the optimization model of the hybrid active neutral point clamped inverter, and T s is the sampling period.

[0091] Those of ordinary skill in the art will realize that the embodiments described herein are for helping the reader understand the principles of the present invention, and it should be understood that the protection scope of the present invention is not limited to such specific statements and embodiments. Those of ordinary skill in the art can make various other specific deformations and combinations without departing from the essence of the present invention according to the technical revelations disclosed in the present invention, and these deformations and combinations are still within the protection scope of the present invention.

Claims

1. An optimization model predictive control method for a hybrid active neutral point clamped inverter, characterized in that: The following steps are involved: S1. Determine a switching state of the optimization model of the hybrid active neutral point clamped inverter according to the topological structure of the optimization model of the hybrid active neutral point clamped inverter; S2. Calculate the evaluation function of the optimization model of the hybrid active midpoint clamped inverter by using the forward Euler method, and determine the candidate virtual voltage vector of the optimization model of the hybrid active midpoint clamped inverter according to the evaluation function of the optimization model of the hybrid active midpoint clamped inverter; Step S2 includes the following steps: S21, calculating the output voltage of the optimization model of the hybrid active neutral point clamped inverter in a two-phase stationary coordinate system; S22, according to the output voltage of the optimization model of the hybrid active neutral point clamped inverter in the two-phase stationary coordinate system in step S21, the output voltage prediction value of the optimization model of the hybrid active neutral point clamped inverter is calculated by using the forward Euler method, which is expressed as: in: For the The output voltage prediction value of the optimization model of the hybrid active neutral point clamped inverter at time , The load inductance of the optimization model for the hybrid active neutral point clamped inverter is is the reference current of the optimization model of the hybrid active neutral point clamped inverter, For the The current sampling value of the optimization model of the hybrid active neutral point clamped inverter at time , is the sampling period, Load resistance for the optimization model of hybrid active neutral point clamped inverter; S23, calculating an evaluation function of the optimization model of the hybrid active neutral point clamped inverter according to the output voltage prediction value of the optimization model of the hybrid active neutral point clamped inverter in step S22; S24, determining a candidate virtual voltage vector of the optimization model of the hybrid active neutral point clamped inverter according to the evaluation function of the optimization model of the hybrid active neutral point clamped inverter in step S23; S3, determining a switching sequence of the optimization model predictive control of the hybrid active neutral point clamped inverter according to the switching state of the optimization model of the hybrid active neutral point clamped inverter in step S1 and the candidate virtual voltage vector of the optimization model of the hybrid active neutral point clamped inverter in step S2; S4. Determine the action time of the optimization model predictive control of the hybrid active neutral point clamped inverter according to the reference voltage of the optimization model of the hybrid active neutral point clamped inverter, and perform predictive control on the optimization model of the hybrid active neutral point clamped inverter in combination with the switching sequence of the optimization model predictive control of the hybrid active neutral point clamped inverter in step S3.

2. The optimization model predictive control method for a hybrid active neutral point clamped inverter according to claim 1, characterized in that: In step S23, the evaluation function of the optimization model of the hybrid active neutral point clamped inverter is calculated, which is expressed as: in: is the evaluation function value of the optimization model of the hybrid active neutral point clamped inverter, is the horizontal coordinate of the output space voltage vector of the optimization model of the hybrid active neutral point clamped inverter, For the The horizontal axis is the output voltage prediction value of the optimization model of the hybrid active neutral point clamped inverter at time, is the output space voltage vector ordinate of the optimization model of the hybrid active neutral point clamped inverter, For the The ordinate is the output voltage prediction value of the optimization model of the hybrid active neutral point clamped inverter at time.

3. The optimization model predictive control method for a hybrid active neutral point clamped inverter according to claim 1, characterized in that: Step S3 includes the following sub-steps: S31, according to the switch state of the optimization model of the hybrid active neutral point clamped inverter in step S1, determine a switch sequence that satisfies the coordinated control of the low frequency unit and the high frequency unit in the optimization model of the hybrid active neutral point clamped inverter; S32, determining a switching sequence that satisfies capacitor voltage balance in the optimization model of the hybrid active neutral point clamped inverter according to the candidate virtual voltage vectors of the optimization model of the hybrid active neutral point clamped inverter in step S2; S33, calculating the output voltage prediction difference of the optimization model of the hybrid active neutral point clamped inverter; S34. Determine a switching sequence for predictive control of the optimization model of the hybrid active neutral point clamped inverter according to the switching sequence that satisfies the coordinated control of the low-frequency unit and the high-frequency unit in the optimization model of the hybrid active neutral point clamped inverter in step S31, the switching sequence that satisfies the capacitor voltage balance in the optimization model of the hybrid active neutral point clamped inverter in step S32, and the output voltage prediction difference of the optimization model of the hybrid active neutral point clamped inverter in step S33.

4. The optimization model predictive control method for a hybrid active neutral point clamped inverter according to claim 3, characterized in that: In step S33, the output voltage prediction difference of the optimization model of the hybrid active neutral point clamped inverter is calculated, which is expressed as: in: For the The DC side capacitor voltage difference of the optimization model of the hybrid active neutral point clamped inverter at the moment, is the DC side capacitance value of the optimization model of the hybrid active neutral point clamped inverter, is a constant, , is the action time of different switching vectors in the switching sequence, , for Phase switch status, for Phase switch status, for Phase switch status, is the transpose symbol, , for Phase current, for Phase current, for Phase current, For the The DC side capacitor voltage difference of the optimization model of the hybrid active neutral point clamped inverter at each moment.

5. The optimization model predictive control method for a hybrid active neutral point clamped inverter according to claim 1, characterized in that: In step S4, the action time of the optimization model predictive control of the hybrid active neutral point clamped inverter is determined according to the reference voltage of the optimization model of the hybrid active neutral point clamped inverter, which is expressed as: in: is the voltage vector The action time, is the first voltage vector in the switching sequence, is the voltage vector The action time, is the second voltage vector in the switching sequence, is the voltage vector The action time, is the third voltage vector in the switching sequence, is the reference voltage of the optimization model of the hybrid active neutral point clamped inverter, is the sampling period.

Images