Discrete space vector modulation low common mode predictive control method for three-level inverter

Abstract

The present invention proposes a low common-mode predictive control method for a three-level inverter using discrete space vector modulation, which can effectively reduce system cost, common-mode voltage, and current ripple. Specifically, it includes: sampling the grid-side voltage and the inverter-side output current, and predicting the voltage value at the next moment based on the current prediction model and deadbeat control; judging the sector in which it is located based on the predicted voltage value and performing sector conversion, and using the conversion result as a reference voltage vector; synthesizing a virtual vector based on discrete space vector modulation, using the three virtual vectors closest to the reference voltage vector as candidate vectors, and the optimal vector is determined by the current tracking value function; selecting the optimal small vector and action time through the midpoint balance value function, and obtaining the optimal vector sequence and action time based on the volt-second balance principle, and equivalently synthesizing the optimal virtual vector; obtaining the switch pulse sequence and action time of the auxiliary bridge arm of the converter and the three-phase half-bridge structure based on the above-mentioned optimal vector sequence and action time, as well as the sector number.

Description

Technical Field

[0001] The present invention belongs to the technical field of three-level eight-switch converters, and in particular relates to a low common mode predictive control method and system for a discrete space vector modulated three-level inverter. Background Art

[0002] The statements in this section merely provide background information related to the present invention and do not necessarily constitute prior art.

[0003] Three-phase multilevel converters are widely used in the industry for medium and high voltage power generation and power transmission. The most mature multilevel topology is the three-level neutral point clamped converter, which uses 12 active power switches and 6 diodes as auxiliary devices. Furthermore, the T-type three-level inverter uses anti-parallel active switches, further saving 6 clamping diodes, and is widely used in photovoltaic power generation.

[0004] The current three-level eight-switch converter system with low common-mode voltage uses only eight active switches and two clamping diodes to achieve three-level output in order to further reduce costs, while having higher reliability due to the reduction of common-mode voltage. However, due to the lack of neutral vectors and the elimination of high common-mode vectors, only 13 voltage vectors are available, which leads to large output current ripple. The inherent midpoint voltage oscillation of the three-level inverter will also increase the voltage stress on the device and increase the ripple in the output current.

[0005] However, traditional modulation algorithms are difficult to directly apply to three-level eight-switch converter systems because, unlike traditional three-level converters, some of the switch tubes of the eight-switch converter are three-phase coupled, resulting in only two levels of three-phase voltage at the same time and missing neutral vector. Summary of the invention

[0006] In order to overcome the deficiencies of the above-mentioned prior art, the present invention provides a low common-mode predictive control method for a three-level inverter with discrete space vector modulation, which can reduce system cost, effectively achieve the reduction of common-mode voltage, control of midpoint potential balance, fixation of switching frequency, fast dynamic response and low current ripple, and significantly improve system performance.

[0007] To achieve the above objectives, one or more embodiments of the present invention provide the following technical solutions:

[0008] In a first aspect, a low common mode predictive control method for a three-level inverter using discrete space vector modulation is disclosed, comprising:

[0009] Sample the grid-side voltage and the three-phase output current of the inverter side, and predict the voltage value at the next moment based on the current prediction model and deadbeat control;

[0010] Based on the predicted voltage value, the sector is determined and sector conversion is performed, and the conversion result is used as a reference voltage vector;

[0011] Synthesize virtual vectors based on discrete space vector modulation, take three virtual vectors closest to the reference voltage vector as candidate vectors, and select the optimal virtual vector through the current tracking value function;

[0012] The optimal small vector and its action time are selected through the midpoint balance value function, and based on the volt-second balance principle, the low common mode vector sequence and action time are obtained to equivalently synthesize the optimal virtual vector;

[0013] Based on the above optimal vector sequence and action time, as well as the sector number, the switching pulse sequence and action time of the auxiliary bridge arm and the three-phase half-bridge structure of the converter are obtained.

[0014] As a further technical solution, the current prediction model is:

[0015]

[0016] Among them, e x is the grid voltage, u x is the output voltage of the inverter, i x is the three-phase output current, L and R are the filter inductor and resistor.

[0017] As a further technical solution, the sampled voltage on the grid side and the sampled current on the inverter side are transformed by coordinates:

[0018]

[0019] Get the sampling value e in the 60 degree coordinate system α (k), e β (k), i g (k), i h (k), and through Lagrange extrapolation, the current value is obtained

[0020] As a further technical solution, according to the deadbeat control, the current prediction model formula is discretized to obtain the voltage value V that makes the current error zero at the next moment. g (k+1) and V h (k+1):

[0021]

[0022] RL filter circuit, R is the filter resistance value, L is the filter inductance value, T s is the sampling time,

[0023] As a further technical solution, the space vector diagram of the inverter is divided into six sectors, each sector occupies 60 degrees, and the sector S where the reference voltage is located is connected via V g (k+1) and V h The positive or negative sign of (k+1) is determined.

[0024] As a further technical solution, the reference voltage is rotated to sector I through sector conversion to simplify the calculation, and the conversion result is used as the reference voltage vector V ref .

[0025] As a further technical solution, based on discrete space vector modulation, the sampling period T s Divide into N equal intervals, and use multiple voltage vectors to fill the preset time intervals to synthesize a virtual voltage vector, which is expressed by the formula:

[0026]

[0027]

[0028] Among them, t j is the actual voltage vector V j The action time is an integer multiple of the preset interval. In this method, the virtual vectors are evenly distributed in the form of triangle vertices, and there are a total of (N+1)(N+2) / 2 virtual vectors.

[0029] As a further solution, the coordinates of the reference voltage vector are divided by V dc / 2 normalized, and then multiplied by N, that is:

[0030]

[0031] The coordinates of the reference voltage vector are then rounded down and represented as h1, h2, and h3 respectively:

[0032]

[0033] When the value of h1+h2 is equal to h3, the reference voltage vector falls in the lower triangle, and the candidate vectors are V1(h1+1,h2), V2(h1,h2+1), and V3(h1,h2); when the value of h1+h2 is equal to h3+1, the reference voltage vector falls in the upper triangle, and the candidate vectors are V1(h1+1,h2), V2(h1,h2+1), and V3(h1+1,h2+1).

[0034] As a further solution, the three candidate vectors are used as variables to select the optimal virtual vector V via the current tracking value function. opt , expressed as:

[0035]

[0036]

[0037] As a further solution, the switching pulse sequence and action time of the auxiliary bridge arm are determined according to the midpoint potential; the switching pulse sequence and action time of the three-phase half-bridge structure are determined according to the midpoint potential and the sector.

[0038] In a second aspect, a low common mode predictive control system of discrete space vector modulation is disclosed, comprising:

[0039] The data prediction module is configured to: sample the grid-side voltage and the inverter-side three-phase output current, and predict the voltage value at the next moment based on the current prediction model and deadbeat control;

[0040] The conversion module is configured to: determine the sector based on the predicted voltage value and perform sector conversion, and the conversion result is used as a reference voltage vector;

[0041] The candidate vector module is configured to: synthesize a virtual vector based on discrete space vector modulation, take three virtual vectors closest to the reference voltage vector as candidate vectors, and select the optimal virtual vector via a current tracking value function;

[0042] The optimal virtual vector module is configured to: select the optimal small vector and its action time through the midpoint balance value function, obtain the optimal vector sequence and action time based on the volt-second balance principle, and equivalently synthesize the optimal virtual vector;

[0043] The control module is configured to obtain the switching pulse sequence and action time of the auxiliary bridge arm of the converter and the three-phase half-bridge structure based on the above-mentioned optimal vector sequence and action time, as well as the sector number.

[0044] One or more of the above technical solutions have the following beneficial effects:

[0045] The present invention uses a three-level inverter low common-mode predictive control method of discrete space vector modulation to reduce common-mode voltage and improve system stability; introduces the concept of virtual vector, uses multi-vector fitting to predict reference voltage, reduces current ripple, and improves the power quality of output current; adopts value function to select the optimal small vector and its action time to achieve midpoint potential balance; cleverly designs the switching sequence to achieve a fixed switching frequency, which is conducive to efficient filtering of current harmonics; the hierarchical algorithm under the 60-degree coordinate system realizes rolling optimization with simple algebraic operations, avoids complex trigonometric function calculations, and greatly reduces calculation pressure.

[0046] Compared with a three-level neutral point clamped converter, the method of the invention reduces the number of power switches by 1 / 3, thereby reducing system costs; the method achieves high reliability and high performance control, is simple to implement, and can be easily extended to a multi-level converter system, with strong practicality; and has broad application prospects in industrial fields such as photovoltaic power generation, electric vehicles, and motor drives.

[0047] Advantages of additional aspects of the present invention will be given in part in the following description, and in part will become obvious from the following description, or will be learned through practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] The accompanying drawings in the specification, which constitute a part of the present invention, are used to provide a further understanding of the present invention. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute improper limitations on the present invention.

[0049] Figure 1 The structure diagram of the three-level eight-switch grid-connected inverter system;

[0050] Figure 2 It is the basic space vector diagram of a three-level eight-switch low common-mode grid-connected inverter system;

[0051] Figure 3 A control block diagram of a three-level eight-switch low common-mode grid-connected inverter system according to the present invention;

[0052] 4(a)-(b) are vector sequences of a three-level eight-switch low common-mode grid-connected inverter system according to the present invention;

[0053] 5(a)-(b) are switching sequences of the auxiliary bridge arm of the three-level eight-switch low common-mode grid-connected inverter system of the present invention;

[0054] 6(a)-(b) are switching sequences of a three-phase half-bridge structure of a three-level eight-switch low common-mode grid-connected inverter system according to the present invention;

[0055] 7(a)-(c) show the DC side capacitor voltage, line voltage, output current and common mode voltage of the three-level eight-switch grid-connected inverter system of the present invention using single vector model predictive control, dual vector model predictive control and discrete space vector modulation low common mode predictive control.

[0056] 8(a)-(c) are fast Fourier transform (FFT) analyses of the three-level eight-switch grid-connected inverter system of the present invention using single-vector model predictive control, dual-vector model predictive control, and low common-mode predictive control of discrete space vector modulation.

[0057] 9(a)-(c) are the dynamic response waveforms of the current of the three-level eight-switch grid-connected inverter system of the present invention using single-vector model predictive control, dual-vector model predictive control, and low common-mode predictive control of discrete space vector modulation. DETAILED DESCRIPTION

[0058] It should be noted that the following detailed descriptions are exemplary and are intended to provide further explanation of the present invention. Unless otherwise specified, all technical and scientific terms used herein have the same meanings as those commonly understood by those skilled in the art to which the present invention belongs.

[0059] It should be noted that the terms used herein are for describing specific embodiments only and are not intended to be limiting of exemplary embodiments according to the present invention.

[0060] In the absence of conflict, the embodiments of the present invention and the features of the embodiments may be combined with each other.

[0061] Embodiment 1

[0062] This embodiment discloses a discrete space vector modulated three-level inverter low common mode predictive control method to achieve control of a three-level eight-switch converter. The method of the present invention is applicable to both DC-AC and AC-DC. The following is a specific description using DC-AC as an example, and further illustrated in conjunction with the accompanying drawings and embodiments.

[0063] In one or more embodiments, the control object of the discrete space vector modulated three-level inverter low common mode predictive control method is a three-level eight-switch converter system, referring to Figure 1 , Figure 1 The topology structure is an eight-switch three-level inverter, which is formed by connecting an auxiliary bridge arm and a three-phase half-bridge structure. The auxiliary bridge arm includes an active switch, two clamping diodes and an active switch connected in series in sequence, and the midpoints of the two clamping diodes are connected to the midpoint of the DC side capacitor; the three-phase half-bridge structure includes three-phase bridge arms connected in parallel, each phase bridge arm includes two active switches connected in series, a total of six active switches, and the midpoint of each phase bridge arm is connected to the AC load via a filter.

[0064] Through different switch combinations, the auxiliary bridge arm generates three levels via a DC voltage source and a capacitor, and one level is selected by switching the upper and lower bridge arms of the three-phase half-bridge structure, which is then used as the output level on the AC side.

[0065] The three levels are: 0. They are denoted as P, O, and N respectively; among them, V dc is the power supply voltage value on the DC side.

[0066] Ideally, the DC power supply V dc The voltage on capacitor C1 is V dc / 2, and the voltage on capacitor C2 is -Vdc / 2. Whether switch S1 is turned on or off determines the voltage V H Yes V dc / 2 is still clamped to zero level via the diode. Similarly, S2 determines V M Equal to -V dc / 2 is still 0. Therefore, the working principle of the auxiliary bridge arm can also be expressed as:

[0067]

[0068] Therefore, the auxiliary bridge arm can generate three levels V dc / 2, 0, -V dc / 2, marked as "P", "O" and "N" respectively. In addition, the auxiliary bridge arm is connected via V H With V M Connected to a three-phase half-bridge structure. Each phase of the three-phase half-bridge structure has two IGBTs, which are connected to the output filter and the load via the midpoint of each phase. The three-phase output voltage is expressed as:

[0069]

[0070] in,

[0071]

[0072] V lo (l = a, b, c) represents the phase voltage. Therefore, according to the conversion formula

[0073]

[0074] Then, 21 voltage vectors of the topology are obtained, which represent the relationship between different switch combinations and the output voltage vectors and are recorded in Table 1.

[0075] Table 1

[0076]

[0077] In order to suppress the common-mode voltage and enhance system stability, the amplitude of the common-mode voltage exceeds V dc The voltage vectors of / 6 are discarded, namely the partially redundant small vectors and zero vectors shown in bold in Table 1. In addition, since the three-phase output voltage of the half-bridge structure is selected between only two levels at a certain moment, there is no middle vector. Therefore, the three-level eight-switch low common-mode inverter system has only 13 available voltage vectors. The active switch is IGBT.

[0078] The common-mode voltage of the three-phase eight-switch converter can be expressed as:

[0079]

[0080] Therefore, the common-mode voltage of the 21 voltage vectors is V dc / 6 amplitude change, the maximum amplitude is V dc / 2, as shown in Table 1.

[0081] Figure 2 The basic space vector diagram of the three-level eight-switch low common-mode grid-connected inverter system. In order to suppress the common-mode voltage and enhance the system stability, the amplitude of the common-mode voltage exceeds V dc The voltage vectors of / 6 are discarded, namely the partially redundant small vectors and zero vectors shown in bold in Table 1. In addition, since the three-phase output voltage of the half-bridge structure is selected between only two levels at a certain moment, there is no middle vector. Therefore, the three-level eight-switch low common-mode inverter system has only 13 available voltage vectors.

[0082] Figure 3 This is a control block diagram of the three-level eight-switch low common-mode grid-connected inverter system of the present invention, which is divided into the steps of prediction model, rolling optimization, switching pulse, etc.

[0083] Prediction model steps: sample the grid-side voltage and the three-phase output current on the inverter side, predict the voltage value at the next moment based on the system's current prediction model and deadbeat control, determine the sector it is in and perform sector conversion, and use the conversion result as the reference voltage vector.

[0084] The current prediction model of the three-phase eight-switch converter system is:

[0085]

[0086] Among them, e x is the grid voltage, u x is the output voltage of the inverter, i x is the three-phase output current, L and R are the filter inductor and resistor. The sampling voltage on the grid side and the sampling current on the inverter side are transformed by coordinates.

[0087]

[0088] Among them, [x a x b x c ] T is the grid voltage e x Or AC side output current i x Coordinates in the abc three-phase coordinate system; [x g x h ] T is the grid voltage e x Or sampling current i x The coordinates in the 60-degree coordinate system are extrapolated by Lagrange to obtain the current value.

[0089] According to the deadbeat control, the discretized current prediction model formula is used to obtain the voltage value V that can make the current error zero at the next moment. g (k+1) and V h (k+1):

[0090]

[0091] Among them, this system uses RL filtering circuit, R is the filter resistance value, L is the filter inductance value, T s is the sampling time.

[0092] The space vector diagram is divided into six sectors, each sector occupies 60 degrees. The reference voltage sector S can be g (k+1) and V h The positive and negative signs of (k+1) are determined as shown in Table 2:

[0093] Table 2

[0094] Sector S I II III IV V VI <![CDATA[V g (k+1)]]> + - - - + + <![CDATA[V h (k+1)]]> + + + - - - <![CDATA[V g (k+1)+V h (k+1)]]> Any + - Any - +

[0095] Furthermore, the sector conversion rotates the reference voltage to sector I to simplify the calculation. The conversion result is used as the reference voltage vector V ref , the coordinates in the 60 degree coordinate system The coordinate transformation is shown in Table 3:

[0096] Table 3

[0097]

[0098] The rolling optimization algorithm is implemented in two layers under the 60-degree coordinate system to improve the computational efficiency: the first layer, based on discrete space vector modulation, can synthesize a large number of virtual vectors, take the three virtual vectors closest to the reference voltage vector as candidate vectors, and select the virtual vector with the best current tracking performance through the value function.

[0099] After the prediction model of the controlled system is established, MPC will solve the optimal control vector again at the arrival of each sampling period, which is called online optimization. The optimization process is an online calculation of the loop body in each control period.

[0100] Based on discrete space vector modulation, the sampling period T s Divide into N equal intervals and fill the preset time intervals with multiple voltage vectors to synthesize a virtual voltage vector, which can be expressed by the formula:

[0101]

[0102]

[0103] Among them, t j is the actual voltage vector V j The action time is an integer multiple of the preset interval. In this method, the virtual vectors are evenly distributed in the form of triangle vertices, and there are a total of (N+1)(N+2) / 2 virtual vectors.

[0104] The coordinates of the reference voltage vector are divided by V dc / 2 normalized, and then multiplied by N, that is:

[0105]

[0106] The coordinates of the reference voltage vector are then rounded down and represented as h1, h2, and h3 respectively:

[0107]

[0108] When the value of h1+h2 is equal to h3, the reference voltage vector falls in the lower triangle, and the candidate vectors are V1(h1+1,h2), V2(h1,h2+1), and V3(h1,h2); when the value of h1+h2 is equal to h3+1, the reference voltage vector falls in the upper triangle, and the candidate vectors are V1(h1+1,h2), V2(h1,h2+1), and V3(h1+1,h2+1). Using the three candidate vectors as variables, the optimal virtual vector V is selected through the current tracking value function. opt , expressed as:

[0109]

[0110]

[0111] The second layer selects the optimal small vector and its action time through the midpoint balance value function. The midpoint balance state is the capacitor voltage value V on the DC side. P Equal to the lower capacitor voltage V N Therefore, in order to achieve midpoint voltage balance, when V P >V N When a P-type small vector is selected, the upper capacitor is discharged, V P Decrease; when V P <V N When N-type small vector is selected, the lower capacitor is discharged, V N The optimal action time of the small vector is based on the midpoint equilibrium value function g v (V j )Sure,

[0112]

[0113] k=min g v(k),k∈Z, (16)

[0114] Where C is the DC side capacitance, T s As the sampling period, T s Divide into N equal intervals, k is the number of time intervals in which the selected small vector acts. o is the midpoint current i of the small vector o , recorded in Table 4 below:

[0115] Table 4

[0116] Vector <![CDATA[V7[POO]]]> <![CDATA[V9[OPO]]]> <![CDATA[V 11 [OPEN]]> <![CDATA[V 14 [ON]]> <![CDATA[V 16 [NOW]]> <![CDATA[V 18 [ONO]]> <![CDATA[i0]]> <![CDATA[i b +i c ]]> <![CDATA[i a +i c ]]> <![CDATA[i a +i b ]]> <![CDATA[-i a -i b ]]> <![CDATA[-i b -i c ]]> <![CDATA[-i a -i c ]]>

[0117] Then, based on the volt-second balance principle, the optimal vector sequence and action time are designed to equivalently synthesize the optimal virtual vector. B1 and V B2 , zero vector V Z and the small vector V S The optimal virtual vector is synthesized by s Divide into N time intervals; the action time is expressed as interval n. The action time of each vector is determined based on the volt-second balance principle:

[0118]

[0119] Furthermore, in the switching pulse part, the switching pulse sequence and action time of the auxiliary bridge arm are determined according to the midpoint balance; the switching pulse and action time of the three-phase half-bridge structure are determined according to the midpoint balance and the sector number.

[0120] Figure 4 (a)-(b) shows the vector sequence of the three-level eight-switch low common-mode grid-connected inverter system of the present invention (taking sector I as an example). The high common-mode vector is discarded, and the amplitude of the common-mode voltage is controlled at V dc / 6, reduced by half

[0121] When V P Greater than V N When , the action vectors are V0, V1, V2 and V7. Based on equations (15)-(17), the action time of the four vectors is:

[0122]

[0123] When V P Less than V N When the action vectors are V0, V1, V2 and V 14 Based on equations (15)-(17), the action time of the four vectors is:

[0124]

[0125] Figure 5 (a)-(b) shows the switching sequence of the auxiliary bridge arm of the three-level eight-switch low common-mode grid-connected inverter system of the present invention. The auxiliary bridge arm determines the vector type and outputs V H and V M , which is also the input of the three-phase half-bridge structure. When S1 is turned on, the V H is P, otherwise it is O; when S2 is turned on, V M is 0, otherwise it is N. Therefore, the level is used to describe the switch state of the auxiliary bridge arm. For example, "PO" means S1 is turned on, S2 is turned off, and V H and V M They are P and O respectively, and the three-phase output level is P or O. Obviously, the auxiliary bridge arm has nothing to do with the sector rotation.

[0126] Therefore, V P Greater than V N When the auxiliary bridge arm is switched on and off, the switching sequence is always: OO-PO-PN-PO-OO, which is divided into five sections. The action time is expressed as:

[0127]

[0128] Therefore, V P Less than V N When the auxiliary bridge arm is on, the switching sequence is: OO-ON-PN-ON-OO, which is divided into five sections. The action time is expressed as:

[0129]

[0130] FIG6 (a)-(b) shows the switching sequence of the three-phase half-bridge structure of the three-level eight-switch low common-mode grid-connected inverter system of the present invention. The three-phase half-bridge structure is a two-level topology structure, which determines the three-phase voltage and is related to the sector rotation. "1" or "0" is used to indicate the switching state of the bridge arm. For example, "100" means that the upper bridge arm of phase A is connected, and the lower bridge arms of phases B and C are connected.

[0131] When V P Greater than V N When the first sector three-phase half-bridge structure sequence is: 100-110-100, the action time is:

[0132]

[0133] The sequence of the three-phase half-bridge structure in other sectors is similar and is summarized in Figure 6(a).

[0134] When V P Less than V N When the first sector three-phase half-bridge structure sequence is: 110-100-110, the action time is:

[0135]

[0136] The sequence of the three-phase half-bridge structure in other sectors is similar and is summarized in Figure 6(b).

[0137] The pulse sequences and action times of all switches are summarized in Table 5.

[0138] Table 5

[0139]

[0140]

[0141] Figure 7 (a)-(c) shows the DC side capacitor voltage, line voltage, output current and common mode voltage of the three-level eight-switch inverter system of the present invention using single vector model predictive control, dual vector model predictive control, and low common mode predictive control of discrete space vector modulation. And the THD of the three methods are 10.44%, 3.76% and 2.97% respectively. Obviously, compared with the existing methods, the proposed method greatly reduces the current ripple and improves the current quality; at the same time, the proposed method reduces the common mode voltage to half of the traditional model predictive control method, improving the system stability; the proposed method selects the appropriate vector action to achieve the balance of the midpoint voltage.

[0142] Figure 8 (a)-(c) is a fast Fourier transform (FFT) analysis of the three-level eight-switch inverter system of the present invention using single vector model predictive control, dual vector model predictive control, and low common mode predictive control of discrete space vector modulation. Compared with the existing method, the current harmonics of the proposed method are smaller and more concentrated at integer multiples of the switching frequency, which is conducive to designing filters to filter out harmonics more efficiently.

[0143] Figure 9 (a)-(c) are the dynamic response waveforms of the current of the three-level eight-switch inverter system of the present invention using single vector model predictive control, dual vector model predictive control, and low common mode predictive control of discrete space vector modulation. It can be seen from the waveforms that the proposed method achieves fast and stable dynamic response while maintaining low current ripple.

[0144] Therefore, the three-level inverter low common mode predictive control method using discrete space vector modulation in the three-level eight-switch converter system of the present invention can reduce the common mode voltage and improve the system stability; introduce the concept of virtual vector, use multi-vector fitting to predict the reference voltage, reduce the current ripple, and improve the power quality of the output current; use the value function to select the optimal small vector and its action time to achieve midpoint potential balance; cleverly design the switch sequence to achieve a fixed switching frequency, which is conducive to the efficient filtering of current harmonics; the hierarchical algorithm under the 60-degree coordinate system realizes rolling optimization with simple algebraic operations, avoids complex trigonometric function calculations, and greatly reduces the calculation pressure. Compared with the three-level midpoint clamped converter, the number of power switches is reduced by 1 / 3, reducing the system cost; the method realizes high reliability and high performance control, is simple to implement, and is easy to expand to multi-level converter systems, with strong practicality; it has broad application prospects in industrial fields such as photovoltaic power generation, electric vehicles and motor drives.

[0145] Embodiment 2

[0146] The purpose of this embodiment is to provide a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the above method when executing the program.

[0147] Embodiment 3

[0148] The purpose of this embodiment is to provide a computer-readable storage medium.

[0149] A computer-readable storage medium stores a computer program, which executes the steps of the above method when executed by a processor.

[0150] Embodiment 4

[0151] The purpose of this embodiment is to provide a low common mode prediction control system of discrete space vector modulation, including:

[0152] The data prediction module is configured to: sample the grid-side voltage and the inverter-side three-phase output current, and predict the voltage value at the next moment based on the current prediction model and deadbeat control;

[0153] The conversion module is configured to: determine the sector based on the predicted voltage value and perform sector conversion, and the conversion result is used as a reference voltage vector;

[0154] The candidate vector module is configured to: synthesize a virtual vector based on discrete space vector modulation, take three virtual vectors closest to the reference voltage vector as candidate vectors, and select the optimal virtual vector via a current tracking value function;

[0155] The optimal virtual vector module is configured to: select the optimal small vector and its action time through the midpoint balance value function, obtain the optimal vector sequence and action time based on the volt-second balance principle, and equivalently synthesize the optimal virtual vector;

[0156] The control module is configured to obtain the switching pulse sequence and action time of the auxiliary bridge arm of the converter and the three-phase half-bridge structure based on the above-mentioned optimal vector sequence and action time, as well as the sector number.

[0157] Compared with the three-level neutral point clamped converter, the present invention reduces the number of power switches by 1 / 3, thereby reducing system cost; reduces common mode voltage, thereby improving system stability; reduces current ripple, thereby improving the power quality of output current; achieves the balance of neutral point voltage and the fixation of switching frequency, thereby facilitating efficient filtering of current harmonics; and replaces complex trigonometric function calculations with simple algebraic operations, thereby reducing calculation pressure.

[0158] The steps involved in the apparatuses of the above embodiments 2, 3 and 4 correspond to the method embodiment 1, and the specific implementation methods can refer to the relevant description part of embodiment 1. The term "computer-readable storage medium" should be understood as a single medium or multiple media including one or more instruction sets; it should also be understood to include any medium that can store, encode or carry an instruction set for execution by a processor and enable the processor to execute any method in the present invention.

[0159] Those skilled in the art should understand that the modules or steps of the present invention described above can be implemented by a general-purpose computer device, or alternatively, they can be implemented by a program code executable by a computing device, so that they can be stored in a storage device and executed by the computing device, or they can be made into individual integrated circuit modules, or multiple modules or steps therein can be made into a single integrated circuit module for implementation. The present invention is not limited to any specific combination of hardware and software.

[0160] Although the above describes the specific implementation mode of the present invention in conjunction with the accompanying drawings, it is not intended to limit the scope of protection of the present invention. Those skilled in the art should understand that various modifications or variations that can be made by those skilled in the art on the basis of the technical solution of the present invention without creative work are still within the scope of protection of the present invention.

Claims

1. A low common mode predictive control method for a three-level inverter using discrete space vector modulation, characterized in that: Applicable to three-level eight-switch converters, including: Sample the grid-side voltage and the three-phase output current of the inverter side, and predict the voltage value at the next moment based on the current prediction model and deadbeat control; According to the deadbeat control, the discretized current prediction model formula is used to obtain the voltage value that makes the current error zero at the next moment. V g (k+1) and V h (k+1): in, is the load resistance value, L is the filter inductance value, is the sampling time; Based on the predicted voltage value, the sector is determined and sector conversion is performed, and the conversion result is used as a reference voltage vector; Synthesize virtual vectors based on discrete space vector modulation, take three virtual vectors closest to the reference voltage vector as candidate vectors, and select the optimal virtual vector through the current tracking value function; The optimal small vector and its action time are selected through the midpoint balance value function, and based on the volt-second balance principle, the low common mode vector sequence and action time are obtained to equivalently synthesize the optimal virtual vector; The coordinates of the reference voltage vector are divided by V dc / 2 normalized, then multiplied by N ,Right now: Then the coordinates of the reference voltage vector are rounded down and expressed as h 1, h 2 and h 3: when h 1+ h The value of 2 is equal to h 3, the reference voltage vector falls in the lower triangle, and the candidate vector is V 1 ( h 1+1, h 2), V 2( h 1, h 2+1), V 3 ( h 1, h 2) When h 1+ h The value of 2 is equal to h When 3+1, the reference voltage vector falls in the upper triangle, and the candidate vector is V 1 ( h 1+1, h 2), V 2 ( h 1, h 2+1), V 3 ( h 1+1, h 2+1); Using three candidate vectors as variables, the optimal virtual vector is selected through the current tracking value function V opt , expressed as: ； Based on the optimal vector sequence and action time, as well as the sector number, the switching pulse sequence and action time of the auxiliary bridge arm and the three-phase half-bridge structure of the converter are obtained.

2. The discrete space vector modulated three-level inverter low common mode predictive control method according to claim 1, characterized in that: The current prediction model is: in, is the grid voltage, is the output voltage of the inverter, is the three-phase output current, and are the filter inductor and load resistor.

3. The discrete space vector modulated three-level inverter low common mode predictive control method according to claim 1, characterized in that: The sampling voltage on the grid side and the sampling current on the inverter side are transformed by coordinates: Get the sampling value in the 60 degree coordinate system e α ( k ), e β ( k ), i g ( k ), i h ( k ), and through Lagrange extrapolation, the current value is obtained .

4. The discrete space vector modulated three-level inverter low common mode predictive control method according to claim 1, characterized in that: The space vector diagram of the inverter is divided into six sectors, each sector occupies 60 degrees, and the sector where the reference voltage is located S via V g (k+1) and V h The positive or negative sign of (k+1) is determined.

5. The discrete space vector modulated three-level inverter low common mode predictive control method according to claim 1, characterized in that: The reference voltage is rotated to sector I through sector conversion to simplify the calculation, and the conversion result is used as the reference voltage vector V ref ; Based on discrete space vector modulation, the sampling period T s Split into N equal intervals, and use multiple voltage vectors to fill the preset time interval to synthesize a virtual voltage vector, which can be expressed as: Among them, t j is the actual voltage vector V j The action time is an integer multiple of the preset interval. In this method, the virtual vectors are evenly distributed in the form of triangle vertices. A virtual vector.

6. A low common mode predictive control system of discrete space vector modulation, using a three-level inverter low common mode predictive control method of discrete space vector modulation as claimed in any one of claims 1 to 5, characterized in that: include: The data prediction module is configured to: sample the grid-side voltage and the inverter-side three-phase output current, and predict the voltage value at the next moment based on the current prediction model and deadbeat control; The conversion module is configured to: determine the sector based on the predicted voltage value and perform sector conversion, and the conversion result is used as a reference voltage vector; The candidate vector module is configured to: synthesize a virtual vector based on discrete space vector modulation, take three virtual vectors closest to the reference voltage vector as candidate vectors, and select the optimal virtual vector via a current tracking value function; The optimal virtual vector module is configured to: select the optimal small vector and its action time through the midpoint balance value function, obtain the optimal vector sequence and action time based on the volt-second balance principle, and equivalently synthesize the optimal virtual vector; The control module is configured to obtain the switching pulse sequence and action time of the auxiliary bridge arm of the converter and the three-phase half-bridge structure based on the above-mentioned optimal vector sequence and action time, as well as the sector number.

7. A computer device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the program, the steps of the method described in any one of claims 1 to 5 are implemented.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the steps of the method described in any one of claims 1 to 5 are performed.

Images