Multi-vector model predictive control method, device and controller with simplified calculation

Abstract

The present invention belongs to the technical field of multi-vector model predictive control and provides a multi-vector model predictive control method and system with simplified calculation. The multi-vector model predictive control method with simplified calculation includes dividing a three-phase space vector diagram into regions to form a plurality of region sets, each region set corresponding one-to-one to a space vector; searching for a plurality of neighboring space vectors that synthesize the reference space vector based on the region set in which the reference space vector is located; calculating all switching state combinations of the converter online based on the converter model, the reference space vector, and all the neighboring space vectors found, with the goal of minimizing the converter's current tracking error, without the need to store any numerical tables such as vector coordinates or state combinations in advance; and selecting the switching state combination with the minimum capacitor voltage deviation to control the converter based on all the calculated switching state combinations of the converter and the voltage control value function, thereby achieving multi-objective optimization control of the fly-by voltage and the midpoint voltage.

Description

Technical Field

[0001] The present invention belongs to the technical field of multi-vector model predictive control, and in particular relates to a multi-vector model predictive control method, device and controller with simplified calculation. 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] Renewable energy sources such as wind and solar are valued for their sustainability and cleanliness. However, their inherent randomness, volatility, and intermittency pose challenges to grid stability. Large-scale grid integration, in particular, can lead to power quality issues such as frequency fluctuations and voltage instability. Multilevel converters, with their advantages for improving power quality and reducing harmonic content, play a key role in integrating renewable energy into the grid.

[0004] Model predictive control (MPC) is widely used in power electronic converters due to its fast dynamic response, simple implementation, and ability to solve nonlinear multi-objective problems. Finite set model predictive control (FCS-MPC), which optimizes within a limited discrete control set, has attracted widespread attention due to its high efficiency and applicability. In particular, it offers greater flexibility by simultaneously considering three-phase voltages and utilizing redundant voltage vectors, such as capacitor voltage balancing and common-mode voltage cancellation.

[0005] In FCS-MPC, the system model is used to predict the future output variables for each switch state. The input that minimizes the cost function is selected as the optimal vector for execution by the converter in the next sampling cycle. However, because traditional FCS-MPC requires numerous prediction calculations in a short period of time, it poses computational challenges to chip processing capabilities. The use of lookup tables to pre-store each voltage vector and switch state requires a certain amount of memory, and the complex space vectors and redundancy of high-level converters impose a significant computational burden. Summary of the Invention

[0006] In order to solve the above technical problems, the present invention provides a multi-vector model predictive control method and system with simplified calculation, which can be applied to all hybrid multi-level converters.

[0007] In order to achieve the above object, the present invention adopts the following technical solutions:

[0008] A first aspect of the present invention provides a multi-vector model predictive control method with simplified calculation.

[0009] In one or more embodiments, a multi-vector model predictive control method with simplified calculation is provided, comprising:

[0010] Divide the three-phase space vector diagram into regions to form several groups of region sets, each group of region sets corresponds to a space vector one-to-one;

[0011] According to the region set where the reference space vector is located, searching for a number of neighboring space vectors for synthesizing the reference space vector;

[0012] Based on the converter model, the reference space vector, and all neighboring space vectors, all switching state combinations of the converter are calculated online with the goal of minimizing the converter's current tracking error. This eliminates the need to store any numerical tables such as vector coordinates or state combinations in advance.

[0013] Based on the calculated switching state combinations and voltage control cost function of the converter, the switching state combination with the minimum capacitor voltage deviation is selected to control the converter to achieve multi-objective optimization control of the flying voltage and midpoint voltage.

[0014] As an implementation manner, a control threshold for controlling the fluctuation range of the flying voltage is preset to determine the priority of controlling the flying voltage and the midpoint voltage, and then the redundant switch state is adjusted according to the priority.

[0015] As an implementation method, the priority relationship between controlling the flying capacitor voltage and the midpoint voltage is:

[0016] The absolute value of the difference between the flying capacitor voltage and the reference value is compared with the control threshold. If the former is greater than the latter, the flying capacitor voltage balance is controlled first, otherwise the midpoint voltage is controlled first.

[0017] As an implementation method, the flying capacitor voltage and the midpoint voltage are balanced and controlled.

[0018] As an implementation method, all predicted switching state combinations of the converter are substituted into the value function of voltage control for comparison. The minimum value function corresponds to the minimum voltage deviation, and the switching state corresponding to the minimum value function is selected to control the operation of the converter's switching tube.

[0019] As an implementation manner, rounding and logical judgment are used to find several neighboring space vectors that synthesize the reference space vector.

[0020] A second aspect of the present invention provides a multi-vector model predictive control system with simplified calculations.

[0021] In one or more embodiments, a multi-vector model predictive control system with simplified computation includes:

[0022] A space vector diagram division module, which is used to divide the three-phase space vector diagram into regions to form several groups of region sets, each group of region sets corresponds to a space vector one-to-one;

[0023] A neighboring space vector search module, which is used to search for a number of neighboring space vectors that synthesize the reference space vector according to the area where the reference space vector is located;

[0024] A switch state combination calculation module is used to calculate all switch state combinations of the converter online based on the converter model, the reference space vector, and all neighboring space vectors found, with the goal of minimizing the converter's current tracking error. This module does not require pre-stored numerical tables such as vector coordinates or state combinations.

[0025] A multi-objective optimization control module is used to select the switching state combination with the minimum capacitor voltage deviation to control the converter based on all the calculated switching state combinations and voltage control value functions of the converter, so as to achieve multi-objective optimization control of the flying voltage and the midpoint voltage.

[0026] As an implementation mode, in the multi-objective optimization control module, a control threshold for controlling the flying voltage fluctuation range is preset to determine the priority of controlling the flying voltage and the midpoint voltage, and then the redundant switch state is adjusted according to the priority.

[0027] As an embodiment, in the multi-objective optimization control module, the priority relationship between controlling the flying capacitor voltage and the midpoint voltage is:

[0028] The absolute value of the difference between the flying capacitor voltage and the reference value is compared with the control threshold. If the former is greater than the latter, the flying capacitor voltage balance is controlled first, otherwise the midpoint voltage is controlled first.

[0029] A third aspect of the present invention provides a controller.

[0030] A controller includes a memory and a computer program stored in the memory and executable on the controller. When the controller executes the program, the steps in the multi-vector model predictive control method with simplified calculation as described above are implemented.

[0031] Compared with the prior art, the present invention has the following beneficial effects:

[0032] (1) The present invention first divides the three-phase space vector diagram into regions, then searches for neighboring space vectors based on the region set where the reference space vector is located, and takes minimizing the current tracking error of the converter as the goal, calculates all the switch state combinations of the converter online, and finally realizes multi-objective optimization control of the fly-through voltage and the midpoint voltage based on the voltage control value function, realizes online calculation of vector coordinates, redundant vectors and duty cycle, avoids the advance input of any numerical table such as vector coordinates or state combinations, eliminates the need to store and read data, and simplifies the calculation of multi-vector model predictive control.

[0033] (2) The present invention sets the control priority of the midpoint voltage and the flying voltage, takes into account the redundant voltage vector, and selects the switch combination with the minimum capacitor voltage deviation to control the converter. This is performed in a three-phase space vector diagram. It can be extended to any multi-level converter by simply correcting the expression between the reference output voltage and the switching function.

[0034] (3) The present invention adopts an implementation method based on region division, which avoids the complex calculation process of traditional MPC. In the current tracking algorithm part, the number of calculations remains unchanged, which is conducive to the promotion and application of multi-level converters.

[0035] (4) The present invention uses simple logical judgment to find several nearest vectors to synthesize the reference vector, so as to minimize the current tracking error. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] The accompanying drawings, 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.

[0037] Figure 1 1 is a system structure diagram of a novel hybrid seven-level converter (T2C-HB) comprising a three-level T-type converter and an H-bridge cascaded in accordance with an embodiment of the present invention;

[0038] Figure 2 is a schematic diagram of area division of a space vector diagram according to one embodiment of the present invention;

[0039] FIG3( a ) is a schematic diagram of a first region division according to an embodiment of the present invention.

[0040] FIG3( b ) is a schematic diagram of a second region division according to an embodiment of the present invention;

[0041] Figure 4 Schematic diagrams of three minimum triangles according to one embodiment of the present invention;

[0042] Figure 5 is a third region division schematic diagram according to one embodiment of the present invention;

[0043] Figure 6 is a schematic diagram of a minimum parallelogram according to one embodiment of the present invention;

[0044] Figure 7 is a third region determination flow chart according to one embodiment of the present invention;

[0045] Figure 8 is a flying capacitor voltage and midpoint voltage control block diagram according to one embodiment of the present invention;

[0046] Figure 9 is an overall control block diagram of a multi-vector MPC calculation method for simplified calculation according to one embodiment of the present invention;

[0047] FIG10( a ) is a three-phase output current simulation diagram according to one embodiment of the present invention;

[0048] FIG10( b ) is a flying capacitor voltage simulation diagram according to one embodiment of the present invention;

[0049] FIG10( c ) is a midpoint voltage simulation diagram according to one embodiment of the present invention;

[0050] FIG10( d ) is a simulation diagram of output line voltage according to one embodiment of the present invention;

[0051] FIG10( e ) is an FFT simulation diagram of the output current according to one embodiment of the present invention;

[0052] Figure 11 This is a flow chart of a multi-vector model predictive control method for simplified calculation according to an embodiment of the present invention;

[0053] Figure 12 It is a schematic structural diagram of a multi-vector model predictive control device with simplified calculation according to an embodiment of the present invention. DETAILED DESCRIPTION

[0054] The present invention will be further described below with reference to the accompanying drawings and embodiments.

[0055] It should be noted that the following detailed descriptions are illustrative and intended to provide further explanation of the present invention. Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the art to which the present invention belongs.

[0056] It should be noted that the terms used herein are only for describing specific embodiments and are not intended to limit the exemplary embodiments according to the present invention. As used herein, unless the context clearly indicates otherwise, the singular form is intended to include the plural form. In addition, it should be understood that when the terms "comprise" and / or "include" are used in this specification, they indicate the presence of features, steps, operations, devices, components and / or combinations thereof.

[0057] Explanation of terms:

[0058] Model predictive control (MPC) is widely used in power electronic converters due to its fast dynamic response, simple implementation, and ability to solve nonlinear multi-objective problems. Among them, finite control set model predictive control (FCS-MPC) has attracted considerable attention due to its high efficiency and applicability. FCS-MPC improves control flexibility by considering redundant voltage vectors to balance capacitor voltages and reduce common-mode voltage. In the FCS-MPC approach, the system model is used to predict the future output variables for each switch state and compare the values ​​of the cost function. However, traditional FCS-MPC requires multiple prediction calculations in a short period of time and uses lookup tables to pre-store each voltage vector and switch state, which consumes a large amount of memory. Furthermore, the many redundant space vectors of high-level converters impose a significant computational burden. Additional constraints, such as midpoint voltage balancing, flying capacitor voltage balancing, and common-mode voltage reduction, further increase the computational requirements of FCS-MPC.

[0059] like Figure 11 As shown, a multi-vector model predictive control method with simplified calculation is provided, including:

[0060] S101: Divide the three-phase space vector diagram into regions to form a plurality of region sets, each region set corresponding to a space vector;

[0061] S102: searching for a number of neighboring space vectors for synthesizing the reference space vector according to the region set where the reference space vector is located;

[0062] S103: Based on the converter model, the reference space vector, and all neighboring space vectors found, all switching state combinations of the converter are calculated online with the goal of minimizing the current tracking error of the converter, without having to store any numerical tables such as vector coordinates or state combinations in advance;

[0063] S104: Based on all the calculated switch state combinations and voltage control cost function of the converter, select the switch state combination with the minimum capacitor voltage deviation to control the converter, so as to achieve multi-objective optimization control of the flying voltage and the midpoint voltage.

[0064] In this embodiment, the number of calculations required for current tracking calculations is independent of the converter's voltage level. The number of computational operations required to find each adjacent vector is fixed, resulting in a constant computational cost and applicability to any multilevel topology. Furthermore, all vector coordinates and switching states in this embodiment are calculated online, eliminating the need for pre-stored numerical tables of vector coordinates or state combinations, further facilitating application to high-level converters. This embodiment is implemented in a three-phase space vector diagram; it can be extended to any multilevel converter by simply modifying the expression between the reference output voltage and the switching function.

[0065] In step S102, rounding and logical analysis are used to search for several neighboring space vectors that form the reference space vector. This avoids the need to compare candidate voltage vectors, reducing computational complexity. The number of computational operations required to search and identify each neighboring vector is predetermined and constant. This means that regardless of the reference vector's position in the space vector diagram, the required computational steps and complexity remain constant, and the algorithm's efficiency and response time do not significantly change with increasing levels.

[0066] In step S103, all vector coordinates and switch states are calculated in real time, eliminating the need to pre-store numerous tables of vector coordinates or switch state combinations. This online calculation significantly reduces memory usage by eliminating the need to allocate additional memory space for each possible vector configuration.

[0067] In step S104, a control threshold for the flying voltage fluctuation range is preset to determine the priority of controlling the flying voltage and the midpoint voltage. The redundant switch states are then adjusted based on the priority. All calculated converter switch state combinations are substituted into the voltage control cost function for comparison. The switch state with the minimum cost function corresponds to the minimum voltage deviation. The switch state corresponding to the minimum cost function is selected to control the converter's switches.

[0068] Among them, the priority relationship between controlling the flying capacitor voltage and the midpoint voltage is:

[0069] The absolute value of the difference between the flying capacitor voltage and the reference value is compared with the control threshold. If the former is greater than the latter, the flying capacitor voltage balance is controlled first, otherwise the midpoint voltage is controlled first.

[0070] In some embodiments, the flying capacitor voltage and the midpoint voltage are balanced. In the ±E state, there are two switch states corresponding to each voltage state, respectively charging or discharging the capacitor. Depending on the positive or negative sign of the two balance functions, the appropriate switch state can be selected to charge or discharge the capacitor.

[0071] Currently, there are three main topologies: midpoint clamped, flying capacitor, and cascaded H-bridge. Each has its own advantages, but also some disadvantages. To balance these drawbacks, hybrid multilevel converters have recently become an area of ​​intensive research. For example, a hybrid cascaded T+H topology utilizes a seven-level converter (T2C-HB) consisting of a three-level T-type and an H-bridge cascade. This not only increases output voltage but also reduces device voltage requirements.

[0072] Since the DC side of the seven-level T2C-HB inverter has two capacitors of the same capacitance connected in series, the voltages of the upper and lower capacitors on the DC side can shift during operation due to numerous factors, leading to a midpoint voltage imbalance. Furthermore, the three-phase flying capacitors of the seven-level T2C-HB inverter also need to be controlled simultaneously. Therefore, in order to achieve multi-objective control of midpoint voltage balance and flying capacitor voltage fluctuation suppression without introducing additional switching losses, this implementation provides a multi-vector model predictive control method that reduces flying capacitor voltage fluctuations, achieves midpoint balance, and reduces computational complexity.

[0073] The control object of this implementation is a new hybrid seven-level converter (T2C-HB) system, such as Figure 1 As shown, each phase consists of a three-level T-type converter and an H-bridge cascade, and the DC side includes two identical series capacitors. A midpoint O is formed between the two capacitors and connected to each phase input of the inverter. Each phase output of the inverter is connected to the grid through a filter. xn ,i x , R, L and U dc They represent the voltage between points x and n, the output current, the resistance, the filter inductance, and the DC side voltage (x=a, b, c).

[0074] Assume that the DC side voltage is U dc (4E), in order to make the seven-level T2C-HB inverter work properly, the DC side capacitor voltage and the flying capacitor voltage should be controlled at U dc / 2 and U dc / 4, i.e., 2E and E. When these conditions are met, the inverter can output seven levels: -3E, -2E, -E, 0, E, 2E, and 3E, corresponding to 12 switching combinations. When the output level is E or -E, two redundant switching combinations exist, respectively, each of which has a charging or discharging effect on the capacitor. Therefore, the flying capacitor voltage and midpoint potential can be controlled by properly allocating the operating time of the two redundant switching combinations.

[0075] First, a discrete mathematical model of the seven-level inverter is established. According to Kirchhoff's law, u xn Can be written as

[0076]

[0077] Using the Euler approximation, (1) in the discrete model is

[0078]

[0079] where x = α, β, u * xn (k) and i * x(k+1) are the reference voltage and reference current at the kth moment and the k+1th moment respectively. x (k) is the output current at the kth moment. s is the sampling period. The reference current i at the k+1th moment * x (k+1), the grid voltage e at the k+1th moment x (k+1) is derived as

[0080] i * x (k+1)=3i * x (k)-3i * x (k-1)+i * x (k-2) (x=a,b,c) (3)

[0081] e x (k+1)=3e x (k)-3e x (k-1)+e x (k-2) (x=α,β) (4)

[0082] like Figure 2 As shown in the figure, in a space vector diagram, any vector is divided into n regions, each of which is called the nth region of the vector. Simply by determining whether the reference vector is located in the nth region, n adjacent vectors can be found without enumeration and iteration, thus reducing the amount of computation.

[0083] In order to simplify the calculation process, the reference vector [V α ,V β ] is normalized to [V α1 ,V β1 ]

[0084]

[0085] The relationship between the conversion between the vector in the mn coordinate system and the vector in the rectangular coordinate system αβ is

[0086]

[0087] Get the coordinates of the vector closest to the reference vector by rounding [m0,n0]

[0088]

[0089] To calculate the first nearest neighbor vector and the second nearest neighbor vector, the first region and the second region are used for judgment. FIG3(a) shows the second region of O2, and FIG3(b) shows the first region of O1.

[0090] Since the reference vector may be located in any area of ​​triangle I or triangle II, it can be judged by the following conditions. The judgment conditions are as follows:

[0091]

[0092] Among them [V m ref ,V n ref ] is the reference vector coordinate in the mn coordinate system. Figure 4 There are three types of triangles in the example. Each vertex of a small triangle consists of two virtual vectors with coordinates (x1, y1) and (x2, y2), and a real vector with coordinates (x3, y3). After finding the vertices of the small triangle, determine whether the vector coordinates satisfy

[0093] V m -V n =3n, n∈N (9)

[0094] A vector that meets this condition is a real vector, otherwise it is a virtual vector. The real vector relative to the first adjacent vector is the second nearest neighbor vector, with coordinates (x1+x2-x3,y1+y2-y3).

[0095] In order to calculate the third nearest neighbor vector, the third area is used to judge, such as Figure 5 As shown, the third area is divided. The relationship between the conversion between the mn60° coordinate system and the rectangular coordinate system αβ is established as follows:

[0096]

[0097] After rounding, a parallelogram area is obtained. Since the reference vector may be located in any area of ​​A, B, C, or D, such as Figure 6 As shown. According to V m ref (k), V n ref (k), m0(k) and n0(k), determine [V m ref ,V n ref ] The boundary conditions of the sub-sector are as follows:

[0098]

[0099] There are four cases as follows, the judgment flow chart is as follows Figure 7 shown.

[0100]

[0101] By analogy, we can determine the nth region where the reference vector is located, and thus find the nth nearest neighbor vector. Since the coordinates and voltage vectors correspond one to one, we can convert the coordinates to obtain all redundant voltage vectors, avoiding the time-consuming table lookup method. The first nearest neighbor vector and the second nearest neighbor vector are converted using the following matrix:

[0102]

[0103] The third nearest neighbor vector is transformed using the following matrix,

[0104]

[0105] Define the base coordinates as [V a ,V b ,V c ]=[V a ,V b ,0], the redundant vector is

[0106] [V a +i,V b +i,i],i∈[n1,n2] (15)

[0107] Where n1=max(-3-V a ,-3-V b ,-3), n2=min(3-V a ,3-V b ,3).

[0108] To calculate the vector duty cycle, taking three vectors as an example, since the three vectors and the reference vector are determined, the cost function can be rewritten as a function of distance:

[0109]

[0110] in Is the reference voltage. Among them, V1, V2 and V3 are used to synthesize V ref The vector of V1, V2 and V3, t1, t2 and t3 are the corresponding duty cycles of V1, V2 and V3. Assume that the value function values ​​of V1, V2 and V3 are J1, J2 and J3 respectively. The longer the distance, the shorter the action time. The duty cycle can be calculated as,

[0111]

[0112] The following equation can be derived:

[0113]

[0114] Since different redundant voltage vectors have different effects on the balance of midpoint and floating capacitor voltages, in order to obtain the optimal switching state, the balance of capacitor voltages can be used as a constraint in the optimization problem, thereby minimizing the error between the expected capacitor voltage and the predicted voltage. Therefore, the voltage balance cost function J2 can be:

[0115]

[0116] The expression of the function BalFC that balances the flying capacitor voltage is:

[0117] BalFC=(U xf -E)i x (x=a,b,c)

[0118] The expression of the equilibrium midpoint voltage function BalNP is:

[0119] BalNP=(U N -U P )i x u xn (x=a,b,c)

[0120] Among them, U xf represents the flying capacitor voltage of phase x (x=a,b,c), U P , U N Respectively represent the voltage of the upper and lower capacitors on the DC side. In the ±E state, there are two switching states corresponding to one voltage output. The appropriate switching state is selected according to the positive or negative of BalFC or BalNP. In order to determine the priority of the flying capacitor voltage and the midpoint voltage control, a threshold is used for selection in the control algorithm. The flow chart is as follows: Figure 8 As shown, by using J2 to obtain the redundant switch state corresponding to the optimal vector, the flying capacitor voltage and the midpoint voltage can be balanced. The overall control flow diagram of the system is shown in Figure 9 As shown in Figure 1, it includes current tracking, midpoint voltage balance control, and flying capacitor voltage control. From the above analysis, it can be seen that the proposed method is performed in a three-phase space vector diagram and can be extended to multi-level converters by simply modifying the relationship between the reference output voltage and the switching function.

[0121] The simulation results are shown in Figure 10. In Figure 10(a), the current is sinusoidal with an amplitude of 6A. In Figure 10(b), the flying capacitor voltage is 1 / 4 of the DC side, with a fluctuation of 0.5V, which is less than Figure 7 The threshold in the figure is 5V; the midpoint voltage in Figure 10(c) is half of the DC side voltage, with a fluctuation of 0.2V; Figure 10(d) is the line voltage; the FFT in Figure 10(e) shows that the THD of the inverter side current is 1.06%, and the current harmonics are effectively suppressed.

[0122] Figure 12 This is a schematic diagram of the structure of a multi-vector model predictive control device for simplified calculation in an embodiment of the present invention. Figure 11 The simplified calculation of the multi-vector model predictive control method corresponds to Figure 12 As shown, the multi-vector model predictive control device with simplified calculation in this embodiment may include:

[0123] A space vector diagram division module 201 is used to divide the three-phase space vector diagram into regions to form a plurality of region sets, each region set corresponding to a space vector;

[0124] A neighboring space vector search module 202 is configured to search for a number of neighboring space vectors for synthesizing the reference space vector according to a set of regions where the reference space vector is located;

[0125] A switch state combination calculation module 203 is used to calculate all switch state combinations of the converter online based on the converter model, the reference space vector, and all neighboring space vectors found, with the goal of minimizing the converter's current tracking error, without requiring the storage of any numerical tables such as vector coordinates or state combinations in advance;

[0126] The multi-objective optimization control module 204 is used to select the switching state combination with the minimum capacitor voltage deviation to control the converter based on all the calculated switching state combinations and voltage control cost function of the converter, so as to achieve multi-objective optimization control of the flying voltage and the midpoint voltage.

[0127] Specifically, in the multi-objective optimization control module 204, a control threshold for controlling the flying voltage fluctuation range is preset to determine the priority of controlling the flying voltage and the midpoint voltage, and then the redundant switch state is adjusted according to the priority.

[0128] In the multi-objective optimization control module 204, the priority relationship between controlling the flying capacitor voltage and the midpoint voltage is:

[0129] The absolute value of the difference between the flying capacitor voltage and the reference value is compared with the control threshold. If the former is greater than the latter, the flying capacitor voltage balance is controlled first, otherwise the midpoint voltage is controlled first.

[0130] It should be noted here that the various modules in the simplified calculation multi-vector model predictive control device correspond one-to-one to the various steps in the simplified calculation multi-vector model predictive control method, and their specific implementation processes are the same and will not be described in detail here.

[0131] In one or more embodiments, a controller is also provided, comprising a memory and a computer program stored in the memory and executable on the controller, wherein when the controller executes the program, the steps in the multi-vector model predictive control method with simplified calculation as described above are implemented.

[0132] In particular, according to an embodiment of the present application, the process described above with reference to the flowchart can be implemented as a computer software program. For example, an embodiment of the present application includes a computer program product, which includes a computer program carried on a computer readable medium, the computer program including a computer program for executing Figure 9 In such an embodiment, the computer program can be downloaded and installed from a network via the communication portion and / or installed from a removable medium. When the computer program is executed by the central processing unit, the various functions defined in the apparatus of the present application are performed.

[0133] in, Figure 9 The computer program instructions corresponding to the method shown can also be stored in a computer readable memory that can guide a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture including an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0134] Those skilled in the art will appreciate that all or part of the processes in the above-described method embodiments can be implemented by instructing related hardware through a computer program. The program can be stored in a computer-readable storage medium, and when executed, the program can include the processes in the above-described method embodiments. The storage medium can be a magnetic disk, an optical disk, a read-only memory (ROM), or a random access memory (RAM).

[0135] The foregoing description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Those skilled in the art will readily appreciate that various modifications and variations of the present invention are possible. Any modifications, equivalent substitutions, or improvements made within the spirit and principles of the present invention are intended to be within the scope of protection of the present invention.

Claims

1. A multi-vector model predictive control method with simplified calculation, characterized in that: include: Divide the three-phase space vector diagram into regions to form several groups of region sets, each group of region sets corresponds to a space vector one-to-one; According to the region set where the reference space vector is located, searching for a number of neighboring space vectors for synthesizing the reference space vector; Based on the converter model, the reference space vector and all the neighboring space vectors, all the switching state combinations of the converter are calculated online with the goal of minimizing the converter's current tracking error. Based on the calculated switching state combinations and voltage control cost function of the converter, the switching state combination with the minimum capacitor voltage deviation is selected to control the converter to achieve multi-objective optimization control of the flying voltage and midpoint voltage.

2. The multi-vector model predictive control method for simplified calculation according to claim 1, characterized in that: A control threshold for controlling the fluctuation range of the flying voltage is preset to determine the priority of controlling the flying voltage and the midpoint voltage, and then the redundant switch state is adjusted according to the priority.

3. The multi-vector model predictive control method for simplified calculation according to claim 2, characterized in that: The priority relationship between controlling the flying capacitor voltage and the midpoint voltage is: The absolute value of the difference between the flying capacitor voltage and the reference value is compared with the control threshold. If the former is greater than the latter, the flying capacitor voltage balance is controlled first, otherwise the midpoint voltage is controlled first.

4. The multi-vector model predictive control method for simplified calculation according to claim 1, characterized in that: The flying capacitor voltage and the midpoint voltage are balanced and controlled.

5. The multi-vector model predictive control method for simplified calculation according to claim 1, characterized in that: All the calculated switching state combinations of the converter are substituted into the value function of the voltage control for comparison. The minimum value function corresponds to the minimum voltage deviation. The switching state corresponding to the minimum value function is selected to control the operation of the converter's switch tube.

6. The multi-vector model predictive control method for simplified calculation according to claim 1, characterized in that: Several neighboring space vectors for synthesizing the reference space vector are found by using rounding and logical judgment.

7. A multi-vector model predictive control device with simplified calculation, characterized in that: include: A space vector diagram division module, which is used to divide the three-phase space vector diagram into regions to form several groups of region sets, each group of region sets corresponds to a space vector one-to-one; A neighboring space vector search module, which is used to search for a number of neighboring space vectors that synthesize the reference space vector according to the region set where the reference space vector is located; A switch state combination calculation module is used to calculate all switch state combinations of the converter online based on the converter model, the reference space vector, and all neighboring space vectors found, with the goal of minimizing the converter's current tracking error; A multi-objective optimization control module is used to select the switching state combination with the minimum capacitor voltage deviation to control the converter based on all the calculated switching state combinations and voltage control value functions of the converter, so as to achieve multi-objective optimization control of the flying voltage and the midpoint voltage.

8. The multi-vector model predictive control device for simplified calculation according to claim 7, characterized in that: In the multi-objective optimization control module, a control threshold for controlling the flying voltage fluctuation range is preset to determine the priority of controlling the flying voltage and the midpoint voltage, and then the redundant switch state is adjusted according to the priority.

9. The multi-vector model predictive control device for simplified calculation according to claim 8, characterized in that: In the multi-objective optimization control module, the priority relationship between controlling the flying capacitor voltage and the midpoint voltage is: The absolute value of the difference between the flying capacitor voltage and the reference value is compared with the control threshold. If the former is greater than the latter, the flying capacitor voltage balance is controlled first, otherwise the midpoint voltage is controlled first.

10. A controller comprising a memory and a computer program stored in the memory and operable on the controller, characterized in that: When the controller executes the program, the steps in the multi-vector model predictive control method with simplified calculation as described in any one of claims 1 to 6 are implemented.

Images