Method and device for controlling three-level inverter and three-level inverter

Abstract

The invention discloses a control method and device of a three-level inverter and the three-level inverter, and can effectively solve the problems that the calculated amount is too large and the real-time control performance is reduced in the current strategy. In the scheme, firstly, based on sampled three-phase voltage, current and reference current, a reference voltage vector under three levels is calculated, the reference vector is converted into an equivalent reference voltage vector under two levels by determining a reference vector, and a simplified control set of the two levels is further obtained; and optimizing and selecting a target control vector through the vector control set. Therefore, the three-level control problem is converted into the two-level control problem, complex sector judgment and trigonometric function operation are avoided, the calculation complexity is remarkably reduced, the execution time of the algorithm is shortened, meanwhile, efficient and stable control is guaranteed, and the method is suitable for large-scale popularization and application. Therefore, the challenge that the calculation demand of the FCS-MPC control strategy is exponentially increased along with the increase of the level number is solved, and the control frequency and the system performance are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of uninterruptible power supply inverter control, and in particular to a control method and device for a three-level inverter and the three-level inverter. Background Art

[0002] Uninterruptible Power Supply (UPS) is widely used to provide stable and uninterrupted power supply to loads, especially when the power grid is abnormal, it can replace the mains to ensure the reliability of power supply and reduce the impact of power grid failure on users. In the back end of the UPS system, the inverter plays a key role, efficiently and stably converting the DC voltage output by the rectifier into high-quality AC voltage, which directly determines the quality of the output voltage at the load end. With the continuous advancement of UPS technology, higher requirements are placed on the design and control strategy of the inverter. The three-level inverter (TLI) has become an important topology structure for inverter design because of its more bridge arm output voltage levels and smaller switch tube voltage stress, which makes its output voltage closer to the standard sine wave, lower harmonic content, and higher voltage quality.

[0003] In traditional technology, the finite set model predictive control (FCS-MPC) method is usually used to control the three-level inverter.

[0004] However, as the number of levels increases, the use of finite set model predictive control methods will consume more computing resources, thus affecting system performance. Summary of the invention

[0005] The purpose of the present invention is to provide a control method and device for a three-level inverter and a three-level inverter, which converts the three-level control problem into a two-level control problem, avoids complex sector judgment and trigonometric function operations, significantly reduces the computational complexity, reduces the execution time of the algorithm, and ensures efficient and stable control, thereby solving the challenge that the computational requirements of the FCS-MPC control strategy increase exponentially with the increase in the number of levels, and improves the control frequency and system performance. In the process of determining the control set, the influence of the midpoint potential difference is considered, and a suitable redundant small vector is selected to achieve midpoint potential balance, avoiding additional hardware circuits or adding weight factors in the value function, and reducing human and material costs.

[0006] In order to solve the above technical problems, the present invention provides a control method for a three-level inverter, comprising:

[0007] Calculate the reference voltage vector of the next control cycle according to the reference current and the three-phase voltage and three-phase current sampled in the current control cycle, wherein the reference voltage vector is a voltage vector under three levels;

[0008] The reference voltage vector is converted into a voltage vector at two levels by using the reference vector to obtain an equivalent reference voltage vector;

[0009] According to the reference vector and the reference voltage vector, an equivalent reference voltage vector is obtained, wherein the equivalent reference voltage vector is a voltage vector at two levels;

[0010] Determine a simplified control set under two levels according to the equivalent reference voltage vector, and convert the simplified control set under two levels into a vector control set under three levels;

[0011] A target control vector is determined according to each control vector in the vector control set, and the three-level inverter is controlled according to the target control vector.

[0012] On the other hand, determining the corresponding reference vector according to the positive and negative signs of the reference voltage vector in the three-phase coordinate system includes:

[0013] Get symbolic function;

[0014] Determine each corresponding sign function value according to the positive and negative signs of the components of each phase of the reference voltage vector in the three-term coordinate system and the sign function;

[0015] The reference vector is determined according to each corresponding sign function value.

[0016] On the other hand, according to the reference vector and the reference voltage vector, an equivalent reference voltage vector is obtained, including:

[0017] The reference voltage vector is subtracted from the reference vector to obtain the equivalent reference voltage vector.

[0018] On the other hand, a simplified control set under two levels is determined according to the equivalent reference voltage vector, including:

[0019] An advance vector and a lag vector of the equivalent reference voltage vector are determined, and a simplified control set under the two levels is determined according to the advance vector and the lag vector.

[0020] On the other hand, determining the leading vector and the lagging vector of the equivalent reference voltage vector, and determining the simplified control set under the two levels according to the leading vector and the lagging vector, including:

[0021] The action time variable is calculated based on the equivalent reference voltage vector;

[0022] Determine the leading vector and lagging vector of the equivalent reference voltage vector and the action time corresponding to the action time variable according to the positive and negative signs of the action time variable;

[0023] The leading vector, the lagging vector and the zero vector are determined as a simplified control set under the two levels.

[0024] On the other hand, the action time variables are calculated based on the equivalent reference voltage vector, including:

[0025] According to the equivalent reference voltage vector The projection on the axis and The projection on the axis and the DC side voltage are used to calculate the first action time variable;

[0026] According to the equivalent reference voltage vector axis and the DC side voltage, and calculating the second action time variable;

[0027] A Z action time variable is calculated according to the first action time variable and the second action time variable.

[0028] On the other hand, determining a target control vector according to each control vector in the vector control set, and controlling the three-level inverter according to the target control vector, comprises:

[0029] comparing the action time of each control vector in the vector control set, and determining the control vector with the largest action time as the target control vector;

[0030] The three-level inverter is controlled according to the target control vector.

[0031] In another aspect, the method further comprises:

[0032] Obtaining a midpoint potential difference sampled in the current control period, and determining a polarity of the midpoint potential difference;

[0033] When the polarity of the midpoint potential difference is positive, determining the redundant small vector as a negative small vector to reduce the midpoint potential difference;

[0034] When the polarity of the midpoint potential difference is negative, the redundant vectorlet is determined to be a positive vectorlet to increase the midpoint potential difference.

[0035] In order to solve the above technical problems, the present invention provides a control device for a three-level inverter, comprising:

[0036] Memory for storing computer programs;

[0037] The processor is used to implement the steps of the control method of the three-level inverter as described above when executing the computer program.

[0038] In order to solve the above technical problem, the present invention provides a three-level inverter, including the control method of the three-level inverter as described above.

[0039] The present invention provides a three-level inverter control method, device and three-level inverter, which can effectively solve the problems of excessive calculation and reduced real-time control performance in current strategies. In this scheme, firstly, based on the reference current and the sampled three-phase voltage and three-phase current, the reference voltage vector under three levels is calculated, and the reference vector is converted into an equivalent reference voltage vector under two levels by determining the reference vector, and a simplified control set of two levels is further obtained; then the target control vector is selected by optimizing the vector control set. It can be seen that the present application converts the three-level control problem into a two-level control problem, avoids complex sector judgment and trigonometric function operations, significantly reduces the computational complexity, reduces the execution time of the algorithm, and ensures efficient and stable control, thereby solving the challenge that the computational requirements of the FCS-MPC control strategy increase exponentially with the increase in the number of levels, and improves the control frequency and system performance. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] In order to more clearly illustrate the embodiments of the present invention, the following briefly introduces the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0041] Figure 1 A flow chart of a three-level inverter control method provided by the present invention;

[0042] Figure 2 An overall control block diagram of a simplified method provided by the present invention;

[0043] Figure 3 A schematic diagram of a two-level voltage vector diagram and sector division provided by the present invention;

[0044] Figure 4 A schematic diagram of a three-level voltage vector diagram and sector division provided by the present invention;

[0045] Figure 5 A schematic diagram of a vector correction provided by the present invention;

[0046] Figure 6 A two-level control set and a voltage action time fast determination algorithm flow chart provided by the present invention;

[0047] Figure 7 A conventional method current THD diagram provided by the present invention;

[0048] Figure 8 A current THD diagram provided by the present invention;

[0049] Fig. 9 A comparison chart of the execution time of a method provided by the present invention and a traditional method in DSP;

[0050] Fig.10 A schematic structural diagram of a three-level inverter control device provided by the present invention. DETAILED DESCRIPTION

[0051] The core of the present invention is to provide a control method and device for a three-level inverter and a three-level inverter, which converts the three-level control problem into a two-level control problem, avoids complex sector judgment and trigonometric function operations, significantly reduces the computational complexity, reduces the execution time of the algorithm, and ensures efficient and stable control, thereby solving the challenge that the computational requirements of the FCS-MPC control strategy increase exponentially with the increase in the number of levels, and improves the control frequency and system performance. In the process of determining the control set, the influence of the midpoint potential difference is considered, and a suitable redundant small vector is selected to achieve midpoint potential balance, avoiding additional hardware circuits or adding weight factors in the value function, and reducing human and material costs.

[0052] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0053] On the one hand, if Figure 1 and Figure 2 As shown, the present invention provides a control method for a three-level inverter, which is applied to a three-level inverter or a processor provided in a three-level inverter. The control method for the three-level inverter includes:

[0054] S11: Calculate the reference voltage vector of the next control cycle according to the reference current and the three-phase voltage and three-phase current sampled in the current control cycle, where the reference voltage vector is a voltage vector under three levels;

[0055] Specifically, the reference current refers to the control current expected to be achieved in the next control cycle, which can be a preset current value. For example, if the output current amplitude of the inverter is expected to be 5A, the reference value can be set to 5A. The sampled data of the three-phase voltage and current represent the current operating state of the system. Among these steps, the core task is to calculate the reference voltage vector required for the next control cycle based on the three-phase voltage, current and reference current sampled in the current control cycle, and the reference voltage vector must be suitable for the working conditions of the three-level inverter. Real-time data is converted or mapped to the corresponding space vector coordinate system to describe the dynamic characteristics of the motor or load. The reference voltage vector is the ideal voltage value required to be output by the inverter in the next control cycle, which indicates how to adjust the inverter switching state to meet the system's requirements for the load voltage. For a three-level inverter, the reference voltage vector not only needs to accurately describe the amplitude and phase of the output voltage, but also needs to consider the characteristics of the three levels, that is, the voltage transition between different levels. This calculation is optimized based on the inverter's control objectives (such as current tracking, power control, or harmonic distortion reduction) to ensure that the voltage waveform output in the next cycle is as close to an ideal sine wave as possible, thereby improving the system's steady-state performance and dynamic response.

[0056] S12: determining a corresponding reference vector according to the positive and negative signs of the reference voltage vector in the three-phase coordinate system;

[0057] In this step, the main task is to determine the corresponding reference vector according to the positive and negative signs of the reference voltage vector calculated in the three-phase coordinate system. This process first resolves the reference voltage vector into three components (for example, Va, Vb, Vc), and determines their signs (i.e., positive or negative) by analyzing the positions of these components in the coordinate system. Here. The reference vector reflects the reference vector under the three-level correction to a reference vector under the two-level correction, and the reference of the subsequent two-level control set to correct back to the three-level vector. This step lays the foundation for the subsequent calculation of the equivalent reference voltage vector, ensuring that the control process of the inverter has a clear goal and direction.

[0058] S13: using the reference vector to convert the reference voltage vector into a voltage vector at two levels to obtain an equivalent reference voltage vector;

[0059] In this step, the main task is to calculate the equivalent reference voltage vector based on the determined reference vector and reference voltage vector, which is used to control the two-level inverter. The principle of this step is to simplify the control calculation complexity by converting the three-level control problem into an equivalent two-level control problem.

[0060] Specifically, the reference vector, as an intermediate variable, represents the current voltage state of the system, while the reference voltage vector is the expected voltage state of the next control cycle. By combining the two, an equivalent reference voltage vector can be determined. Specifically, in this embodiment, by subtracting the reference vector from the reference voltage vector, an equivalent reference voltage vector under two levels can be obtained, which is convenient for calculating the simplified control set under the equivalent reference voltage vector, and then for determining the vector control set under the two-level state, and further determining the optimal target control vector from the vector control set. The step of simplifying the control set is performed under two levels, which can simplify the calculation process and avoid complex operations under three levels, but the target control vector of three levels can still be obtained in the end.

[0061] according to Figure 4 The three-level voltage vector diagram can be divided into 6 large sectors from I to IV. Each large sector can be divided into 6 small sectors marked by hexagonal boxes. These 6 small sectors are Figure 5 The left half of the figure assumes that the reference voltage vector is located in the large sector I. At this time, the six small sectors closest to the reference voltage vector are converted into standard two-level vectors (i.e. Figure 5 On the right side of the reference voltage vector, it is necessary to subtract the reference vector v1(0-1-1), and when the reference voltage vector is located in different large sectors, it is necessary to subtract the corresponding reference vector.

[0062] Since the switching state of two-level control is simpler and the computational burden is lower, the three-level voltage problem is simplified to a two-level equivalent voltage vector, which not only reduces the complexity of the FCS-MPC control algorithm, but also ensures the feasibility and stability of real-time control. This step lays the foundation for the subsequent determination of the simplified control set, making the entire control strategy more efficient while taking into account accuracy.

[0063] S14: determining a simplified control set under two levels according to the equivalent reference voltage vector, and converting the simplified control set under two levels into a vector control set under three levels;

[0064] In this step, the core task is to determine the simplified control set under two levels based on the calculated equivalent reference voltage vector, and convert the simplified control set into a vector control set under three levels. This process first starts from the equivalent reference voltage vector and identifies the voltage output that can be achieved under two levels, which usually involves analyzing the available switch states to find out which switch combinations can most effectively generate the desired output voltage. By establishing a simplified control set, the optimal switch state can be quickly found within the two-level control framework to reduce the computational complexity. Then, this simplified control set needs to be mapped back to the three-level control framework to ensure that these switch states can actually be executed in the three-level inverter. This mapping process is key because it ensures that the computational efficiency is achieved while maintaining the output quality and performance of the three-level inverter.

[0065] This method can not only effectively reduce the real-time computing burden of the control algorithm, but also improve the system response speed and control accuracy, so that the inverter can better adapt to load changes and power demand. This step ensures the consistency and efficiency of the entire control strategy, and lays a solid foundation for the subsequent determination of the target control vector.

[0066] S15: determining a target control vector according to each control vector in the vector control set, and controlling the three-level inverter according to the target control vector.

[0067] In this step, the main task is to determine the target control vector according to each control vector in the vector control set determined in the previous step, and use the target control vector to actually control the three-level inverter. This process first requires evaluating all possible control vectors in the vector control set to select the best control vector that best meets the current load demand and control target. In a preferred embodiment, the target control vector is the voltage vector with the longest action time (the action time is inversely proportional to the value function. When the value function is the smallest, the actual voltage vector output by the three-level inverter is closest to the ideal reference voltage vector. Therefore, the target voltage vector with the longest action time is selected as the target voltage vector, which is conducive to the control of the three-level inverter). In this process, the target control vector not only needs to meet the instantaneous requirements of voltage and current, but also must take into account the dynamic response and steady-state performance of the system to ensure that the inverter can operate reliably under various working conditions. Once the target control vector is determined, the switching elements of the inverter will be controlled through corresponding control strategies, such as PWM (pulse width modulation) technology, to achieve the required voltage output.

[0068] This link is the key to the entire control strategy, which directly affects the output quality and response speed of the inverter, ensuring that the inverter can provide power efficiently and stably in a complex power environment, thereby meeting the user's high standards for power quality.

[0069] In summary, the three-level inverter control method provided by the present invention calculates the reference voltage vector under three levels based on the sampled three-phase voltage, current and reference current, and converts it into an equivalent reference voltage vector under two levels by determining the reference vector, and further obtains a simplified control set of two levels; and then optimizes and selects the target control vector through the vector control set. It can be seen that the present invention converts the three-level control problem into a two-level control problem, avoids complex sector judgment and trigonometric function operations, significantly reduces the computational complexity, reduces the execution time of the algorithm, and ensures efficient and stable control, thereby solving the challenge that the computational requirements of the FCS-MPC control strategy increase exponentially with the increase in the number of levels, and improves the control frequency and system performance.

[0070] On the basis of the above embodiments: as a preferred embodiment, the corresponding reference vector is determined according to the positive and negative signs of the reference voltage vector in the three-phase coordinate system, including: obtaining a sign function; determining each corresponding sign function value according to the positive and negative signs of the components of the reference voltage vector in the three-phase coordinate system, and the sign function; determining the reference vector according to each corresponding sign function value.

[0071] In a specific embodiment, the symbolic function may be predefined, such as the expression of the symbolic function is: ; The expression of the reference vector is: ;sig(u0 x ) is the symbol function, u x is the component of the reference voltage vector in phase x, abc represents phases ABC respectively, v base is the base vector.

[0072] In this preferred embodiment, the determination of the reference vector depends on the analysis of the positive and negative signs of the three-phase voltages in the three-phase coordinate system. First, a sign function is defined to determine the sign state of each reference phase voltage. The specific expression is: x When it is greater than zero, sig(u x ) returns 0; when u x When it is less than zero, sig(u x ) returns -1, where x represents the different phases (A, B, C). By comparing the three-phase components u of the reference voltage vector a 、u b and u c Applying this symbol function, we can get the symbol states corresponding to A, B, and C respectively. These symbol states are combined into a three-dimensional vector, namely the reference vector v base =(sig(u a ),sig(u b ),sig(u c )).

[0073] The reference vector not only provides key information for subsequent control decisions, but also simplifies the complexity of the control algorithm, allowing the corresponding switch state to be determined more quickly during the control process. Through this method, the system can effectively reflect the real-time characteristics of the voltage, ensuring that the inverter can make appropriate control responses under different working conditions, thereby improving the quality and stability of the output voltage.

[0074] As a preferred embodiment, obtaining an equivalent reference voltage vector according to a reference vector and a reference voltage vector includes: subtracting the reference vector from the reference voltage vector to obtain the equivalent reference voltage vector.

[0075] In this preferred embodiment, the generation process of the equivalent reference voltage vector is calculated based on the difference between the reference vector and the reference voltage vector. Specifically, it is necessary to first clarify the voltage state represented by the reference vector, which is determined by the sign function and reflects the current voltage polarity of the system. Then, the reference voltage vector is regarded as the desired voltage output, which is the ideal value that the inverter needs to achieve in the next control cycle. By subtracting the reference vector from the reference voltage vector, the equivalent reference voltage vector can be obtained, for example, Figure 4 The six small sectors in Figure 5 The left half of the figure assumes that the reference voltage vector is located in the large sector I. At this time, the six small sectors closest to the reference voltage vector are converted into standard two-level vectors (i.e. Figure 5 When the reference voltage vector is located in different large sectors, the corresponding reference vector needs to be subtracted to obtain the equivalent reference voltage vector under two levels. When the simplified control set is calculated based on this equivalent reference voltage vector, the calculation process can be simplified to avoid complex sector judgment and trigonometric function operations.

[0076] It can be seen that the calculation of the equivalent reference voltage vector simplifies the subsequent control process because it transforms the complex three-level control problem into a simpler two-level control framework, which helps to improve the real-time and effectiveness of the control algorithm and ensure that the inverter can provide high-quality power output under different load conditions.

[0077] As a preferred embodiment, determining a simplified control set at two levels according to an equivalent reference voltage vector includes: determining a leading vector and a lagging vector of the equivalent reference voltage vector, and determining a simplified control set at two levels according to the leading vector and the lagging vector.

[0078] In this preferred embodiment, the process of determining the simplified control set under two levels depends on the analysis of the equivalent reference voltage vector, specifically including identifying its leading vector and lagging vector. In space vector pulse width modulation (SVPWM) control, the leading and lagging vectors refer to the two adjacent basic vectors of the reference voltage vector in the vector synthesis process, the leading vector leads the reference voltage vector, and the lagging vector lags the reference voltage vector. SVPWM control controls the output voltage of the three-phase inverter by synthesizing a reference vector, thereby realizing the control of the inverter. The reference voltage vector is the ideal voltage vector that the inverter is expected to output as described in the above embodiment, which is calculated according to the control requirements of the actual load. The calculation of these two vectors helps to evaluate the effective control strategy required under given voltage conditions. By combining these two vectors, the control system can form a simplified control set containing possible switch states, which contains the optimal switch combination required to achieve the desired output voltage. By focusing on the leading and lagging vectors, the control switch state can be selected more efficiently, thereby achieving fast response and low computational complexity.

[0079] This method effectively reduces the computational complexity of the control algorithm, while ensuring that the inverter can maintain stable and efficient power output under varying load conditions, improving the overall control performance.

[0080] As a preferred embodiment, a leading vector and a lagging vector of an equivalent reference voltage vector are determined, and a simplified control set under two levels is determined according to the leading vector and the lagging vector, including: calculating an action time variable based on the equivalent reference voltage vector; determining the leading vector and the lagging vector of the equivalent reference voltage vector and their corresponding action times according to the positive and negative signs of the action time variable; and determining the leading vector, the lagging vector and the zero vector as a simplified control set under two levels.

[0081] In this preferred embodiment, if the number of action time variables is 3, namely X, Y, and Z, the process of determining the leading vector and lagging vector of the equivalent reference voltage vector involves the calculation of the three action time variables X, Y, and Z, which are obtained based on the specific value of the equivalent reference voltage vector. First, by analyzing the equivalent reference voltage vector, the three action time variables can be derived, which respectively represent the control time length under different switching states. The positive and negative signs of XYZ reflect the information of the small sector where the equivalent reference vector is located. When the small sector is determined, the corresponding leading and lagging vectors can be determined. Therefore, the leading and lagging vectors can be quickly determined by the positive and negative signs of XYZ. Next, according to the signs of these action time variables, the corresponding leading vector and lagging vector can be effectively distinguished, and the action time required for them in the current control cycle can be clarified. Finally, the leading vector, the lagging vector, and the zero vector are grouped into a simplified control set under two levels.

[0082] This control set not only simplifies the switching state selection process of the inverter, but also ensures that the system can respond quickly when performing voltage control, improves the dynamic performance and stability of the inverter, while reducing the calculation complexity, providing support for efficient and accurate power output.

[0083] As a preferred embodiment, the action time variable is calculated based on the equivalent reference voltage vector, including: The projection on the axis and The projection on the axis and the DC side voltage are used to calculate the first action time variable; according to the equivalent reference voltage vector The projection on the axis and the DC side voltage are used to calculate the second action time variable; and the third action time variable is calculated according to the first action time variable and the second action time variable.

[0084] In a preferred embodiment, the first action time variable is X, the second action time variable is Y, and the third action time variable is Z. The expressions of the three action time variables are: ; Among them, XYZ are three action time variables, , The equivalent reference voltage vector is Axis and Projection on the axis, u dc is the DC side voltage.

[0085] In this preferred embodiment, the expressions of the three action time variables are defined by the specific values ​​of the equivalent reference voltage vector to provide the necessary time parameters for the control of the inverter. and Respectively represent the equivalent reference voltage vector Axis and The projection on the axis, and U dc is the DC side voltage.

[0086] Through these three action time variables, the control system can clearly understand the impact of each voltage state on the inverter operation, so as to accurately adjust the switch state in the actual control process and achieve efficient control of the output voltage. This control method based on action time variables provides strong support for the efficient operation of the inverter in a complex power environment.

[0087] As a preferred embodiment, determining a target control vector according to each control vector in a vector control set, and controlling the three-level inverter according to the target control vector includes:

[0088] Compare the action time of each control vector in the vector control set, and determine the control vector with the largest action time as the target control vector;

[0089] The three-level inverter is controlled according to a target control vector.

[0090] In this preferred embodiment, the target control vector is determined by comparing the action time of each control vector in the vector control set, with the purpose of selecting the vector that makes the output value closest to the reference voltage vector.

[0091] Specifically, the action time is negatively correlated with the value function, while the value function is positively correlated with the difference between the reference voltage vector and the actual output value. That is, if you want the actual output value to be closer to the ideal reference voltage vector, you need to select the target control vector that minimizes the value function. Since the action time is negatively correlated with the value function, you need to select the control vector with the largest action time as the target control vector to make the actual output value closest to the reference voltage vector. Among them, each control vector represents the switching state of the inverter within a specific time, and the action time indicates the length of time that the state is maintained. By analyzing these action times, it is possible to identify which control vector has the longest action time under given conditions, and this vector is therefore selected as the target control vector, so as to more effectively adjust the output voltage to achieve the desired control effect. Finally, based on the selected target control vector, the control system will issue corresponding instructions to the three-level inverter to achieve efficient and stable voltage conversion.

[0092] This approach optimizes the inverter's response speed and overall performance by focusing resources on the most important control vectors, thereby improving the operating efficiency of the power system.

[0093] As a preferred embodiment, the method also includes: obtaining the midpoint potential difference sampled in the current control period, and determining the polarity of the midpoint potential difference; when the polarity of the midpoint potential difference is positive, determining the redundant small vector as a negative small vector to reduce the midpoint potential difference; when the polarity of the midpoint potential difference is negative, determining the redundant small vector as a positive small vector to increase the midpoint potential difference.

[0094] In the prior art, in the FCS-MPC application of three-level inverters, the value function not only needs to ensure the accuracy of the converter output current, but also needs to add additional weight coefficients to adjust the midpoint potential to ensure the balance of the DC side capacitor voltage (the offset of the midpoint potential will cause the two DC side capacitors to be unevenly voltageed, making one of the capacitors vulnerable to damage due to excessive voltage; in addition, it will also cause changes in the three-level vector diagram, resulting in distortion of the inverter output voltage; the fluctuation of the midpoint potential will also introduce low-order harmonics, seriously reducing the power quality). However, adjusting the weight coefficients to achieve the global optimal solution is a complex task that requires a trade-off between different control objectives. At present, the adjustment of the weight coefficients mainly relies on experience and lacks theoretical guidance, which increases the difficulty of controller design and implementation.

[0095] In order to solve this problem, the preferred embodiment introduces a new method, which can dynamically adjust the redundant small vector by obtaining the midpoint potential difference of the current control cycle and determining its polarity. The midpoint potential difference refers to the potential difference of the inverter output voltage relative to the midpoint. By determining its polarity (positive or negative), the system can effectively determine the state of the output voltage.

[0096] When the polarity of the midpoint potential difference is positive, it means that the output voltage is higher than the midpoint potential. At this time, a negative small vector is introduced as a redundant small vector to reduce the midpoint potential difference and prevent the output voltage from being too high, thereby ensuring the safety and stable operation of the equipment; on the contrary, when the polarity of the midpoint potential difference is negative, it means that the output voltage is lower than the midpoint potential. At this time, a positive small vector is used as a redundant small vector to increase the midpoint potential difference and thus enhance the output voltage.

[0097] This dynamic adjustment mechanism based on midpoint potential difference replaces the traditional weight coefficient adjustment, which not only simplifies the control algorithm, but also effectively ensures the balance of the DC side capacitor voltage, improves the stability and control accuracy of the system, and reduces the risk of instability caused by voltage waveform distortion. Through this method, it can respond to voltage changes in real time, optimize the control effect, and enhance the adaptability of the inverter under different load conditions.

[0098] In a specific embodiment, the control method of the three-level inverter includes the following steps:

[0099] First, the system samples the three-phase voltage, three-phase current and midpoint voltage signals of the current control cycle, and calculates the reference voltage vector of the next control cycle based on the deadbeat control principle.

[0100] Specifically, the vector space is partitioned, such as Figure 4 As shown, with 6 redundant small vectors as the center, it is divided into 6 large sectors, specifically: Figure 3As shown, starting from -30°, the entire space vector is divided into 6 sectors at intervals of 60°. It is worth mentioning that the sector division here is only an auxiliary function, and the entire algorithm does not involve complex sector judgment, and the algorithm complexity will not increase as a result.

[0101] According to the idea of ​​deadbeat control, when the reference voltage vector acts for one cycle, the output current is equal to the reference current, and the reference voltage vector of the next control cycle can be obtained by reverse calculation. .

[0102] Second, the reference voltage vector is corrected. First, the signs of the three-phase components of the reference voltage vector are determined to determine the reference vector; then, the reference voltage vector is subtracted from the reference vector to obtain an equivalent reference voltage vector under two levels.

[0103] Specifically, the reference vector is determined as follows:

[0104] like Figure 5 As shown, taking the first sector as an example, considering that the difference between the voltage vector in each large sector and the standard two-level voltage vector is the negative small vector in the corresponding large sector, the reference vector v base Set to the negative small vector among the redundant small vectors closest to the reference voltage vector.

[0105] Taking the first sector as an example, the reference vector selected at this time is (0-1-1), and the signs of the phase components of the reference voltage vector in the abc three-phase coordinate system are (+--), that is, .definition:

[0106] ,;

[0107] Then the expression of the reference vector is:

[0108] , ;

[0109] Without loss of generality, the same results can be obtained for other sectors. Therefore, only two formulas, namely formula (1) and (2), are needed to quickly obtain the reference vector, thus avoiding complicated sector judgment and trigonometric function calculation.

[0110] After the reference vector is obtained, the reference voltage vector is subtracted from the reference vector to obtain a two-level equivalent reference voltage vector, thereby completing the correction of the reference voltage vector and using it as a basis for simplifying the control set judgment.

[0111] Third, according to the equivalent reference voltage vector, three variables about the action time are defined, as shown in formula (5). Then, based on the idea of ​​SVPWM, the simplified control set under two levels and the action time of each vector are quickly determined according to the positive and negative properties of the three variables XYZ.

[0112] Specifically, the three voltage vectors closest to the reference vector are quickly determined according to the effective voltage action time of the traditional two-level SVPWM, and used as the simplified control set of the model predictive control, specifically:

[0113] a) If Figure 3 As shown in the two-level SVPWM strategy, according to the nearest three-vector principle, the voltage vector is divided into 6 sectors, each of which contains a zero vector v0 and two effective voltage vectors v x and v y In one switching cycle, the action vector can be expressed as .Will As a simplified control set of model predictive control, it must contain the zero vector, so only two effective voltage vectors need to be determined to determine the simplified control set;

[0114] b) For two-level SVPWM modulation, the switching state sequence within one switching cycle is: v0-v x -v y -v7-v y -v x -v0, (3);

[0115] Formula (3) is for odd-numbered sectors. For even-numbered sectors, just change v x 、v y Just reverse the order. x and v y Corresponding to v * Lead and lag vectors. Combined Figure 5 The two-level vector diagram in the figure shows that v y {-v a (011),-v b (101),-v c (110)}, and determine v x and v y The simplified control set is determined.

[0116] c) For v x and v y To determine, define three variables and quickly judge according to their signs, as follows:

[0117] According to the vector calculation principle: "a vector can be expressed as a linear combination of any two non-collinear unit vectors", the basic vector v is selected a (100) and v b (010) as these two non-collinear unit vectors, the equivalent reference voltage vector can be expressed as: * =X·va +Y·v b , (4);

[0118] Because v a and v b exist The expressions for the axes are , so we get:

[0119] , ;

[0120] According to the volt-second balance principle, we know that XY is the action time of the voltage vector (100) and (010) in the current control cycle, but the selection of (100) and (010) does not conform to the principle of the nearest three vectors, so further processing is required. Combined with the principle of two-level SVPWM control, that is:

[0121] v * =T x v x +T y v y , ;

[0122] Taking the first sector as an example, at this time v x and v y With v a and -v c , where -v c =v a +v b , so we can get:

[0123] v a ·T x +(v a +v b )·T y =X·v a +Y·v; (7);

[0124] Solve for T x =XY,T y =Y. Let ZXY, similarly, the relationship between the action time of the basic vector closest to the reference vector in each sector and XYZ can be obtained.

[0125] The positive and negative signs of XYZ reflect the information of the small sector where the equivalent reference vector is located. Once the small sector is determined, the corresponding leading vector and lagging vector can be determined. Therefore, the leading and lagging vectors can be quickly determined by the positive and negative signs of XYZ. Figure 3 , Figure 5 Analysis shows that when the reference vector is located in sector A, the leading vector v x =v a, which satisfies ; Lag vector v y Include -v b and -v c Two voltage vectors, when Y>0, v y =-v c , on the contrary v y =-v b Similar results can be obtained for other sectors. Therefore, v can be quickly determined by simply judging the positive and negative of XYZ. x and v y .

[0126] For further analysis, let's take sector A as an example. At this time, v y The action time T y =sign(Y)·Y, other sectors have similar results. Therefore, after the leading vector is determined, the action time of the lagging vector can also be quickly determined.

[0127] For the calculation of the action time of the leading vector, combined with formula (6), we get: T z =1-T x -T y , (7);

[0128] It is worth mentioning that the action time here is in the form of duty cycle, which means the ratio of action time to control period. In addition, to prevent over-modulation, when T x +T y >1 o'clock, order .

[0129] Through the above simple calculation and symbol judgment, the simplified control set and the action time of the three voltage vectors in the control set can be quickly determined. Figure 6 .

[0130] Fourth, after obtaining the simplified control set under two levels, it cannot be directly applied to the model predictive control algorithm, and the simplified control set needs to be reversed. The vectors in the simplified control set are added with the reference vector and restored to the three-level voltage vector as the input of the model predictive control.

[0131] Specifically, after obtaining the simplified control set, the voltage vector at this time is two-level and cannot be used directly, and must be restored to three-level, as follows: , (8).

[0132] Fifth, the premise of stable operation of the three-level converter is to ensure the balance of the midpoint potential. Therefore, taking current control as an example, in addition to the most important current error term, the value function also needs to add the midpoint potential difference term. However, the units of the current error and the midpoint potential difference are inconsistent, and an additional weight factor needs to be introduced to adjust the balance between the two to achieve the ideal control effect. However, in practice, the adjustment of the weight factor is very dependent on the experience of engineers and is time-consuming and labor-intensive. Therefore, the method provided by the present invention can achieve the balance of the midpoint potential and avoid the use of the weight factor through a simple judgment. Specifically, the polarity of the midpoint potential difference is judged. When the midpoint potential difference is positive, the redundant small vector selects a negative small vector to charge the lower bridge arm capacitor and reduce the midpoint potential difference. Therefore, the redundant small vector remains unchanged as the reference vector; when the midpoint potential difference is negative, the redundant small vector selects a positive small vector to charge the upper bridge arm capacitor and increase the midpoint potential difference. At this time, the redundant small vector becomes v base +(111).

[0133] Sixth, according to the simplified control set and action time obtained in the above steps, the voltage vector with the largest action time is selected as the optimal vector output, thereby realizing the control of the three-level converter.

[0134] The method provided by the present invention can greatly reduce the amount of calculation of the model predictive control of the three-level converter, convert the three-level into two-level through vector correction, quickly determine the simplified control set under the two-level, simplify the control set to three voltage vectors, and reduce the time of the iterative operation of the model predictive control. The simplified calculation method designed by the present invention converts the three-level into two-level through vector correction, and designs an algorithm to quickly determine the two-level control set, thereby avoiding the judgment of the sector and the complex trigonometric function calculation. In the process of determining the control set, the influence of the midpoint potential difference is considered, and the appropriate redundant small vector is selected to achieve the midpoint potential balance, thereby avoiding additional hardware circuits or adding weight factors to the value function, and reducing the cost of manpower and material resources. In addition, the method provided by the present invention ensures that the voltage vector output in each control cycle is optimal through strict restriction conditions, thereby ensuring that the control performance of the system is optimal.

[0135] Figure 7 and Figure 8 3 is a comparison diagram of the current THD of the method provided by the present invention and the conventional method. From the comparison results, it can be seen that after adopting the method of the present invention, the current THD is lower and the control performance is improved.

[0136] Combination Figure 7 , Figure 8 and Fig. 9By comparative analysis, it can be seen that the algorithm provided by the present invention can reduce the execution time of the algorithm to a certain extent and improve the output voltage quality on the basis of ensuring the current THD. Therefore, it can also be explained that the simplified calculation method of the three-level converter model predictive control provided by the present invention can effectively reduce the complexity of the algorithm and the rationality of the provided method.

[0137] In addition, it should be noted that the method provided by the present invention can also be applied to inverters of higher levels, taking a five-level inverter as an example, including but not limited to a five-level inverter. Similarly, by converting the five-level voltage vector into a three-level through vector correction, the method provided by the present invention can be used to quickly simplify the control set.

[0138] Second, as Fig.10 As shown, the present invention provides a control device for a three-level inverter, comprising: a memory for storing a computer program; a processor for implementing the steps of the control method for the three-level inverter as described above when executing the computer program.

[0139] In addition, the control device of the three-level inverter may further include external devices, including one or more devices (such as a keyboard, a touch screen, etc.), which can communicate interactively with the controller through an I / O interface.

[0140] In the method provided by the present invention, the processor is selected from DSP (Digital Signal Processing) + FPGA (Field-Programmable Gate Array). DSP has the characteristics of high processing accuracy, good stability, strong reliability, and fast processing speed. It has higher performance and efficiency when performing digital signal processing. Using DSP as a processor can improve the controller's control accuracy of the output current and midpoint potential of the three-level inverter, and can also ensure the rapidity and stability of the current control process. FPGA has the ability of parallel communication and can quickly realize the collection of signals such as voltage and current. In addition, FPGA is also responsible for the generation of switch tube drive signals. The use of DSP+FPGA processing system can give full play to the advantages of both, and the low latency and high-performance data transmission function between the processor and programmable logic can maximize the characteristics of the model predictive control algorithm.

[0141] For other introductions to the control device of the three-level inverter, please refer to the above embodiments, which will not be described in detail in this application.

[0142] In a third aspect, the present invention provides a three-level inverter, including the control method of the three-level inverter as described above. For other introductions to the three-level inverter, please refer to the above embodiments, which will not be described in detail in this application.

[0143] It should also be noted that, in this specification, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "comprise", "include" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, an element defined by the statement "comprises a ..." does not exclude the presence of other identical elements in the process, method, article or device including the element.

[0144] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present invention. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but rather to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A control method for a three-level inverter, characterized in that: include: Calculate the reference voltage vector of the next control cycle according to the reference current and the three-phase voltage and three-phase current sampled in the current control cycle, wherein the reference voltage vector is a voltage vector under three levels; Determine a corresponding reference vector according to the positive and negative signs of the reference voltage vector in the three-phase coordinate system; The reference voltage vector is converted into a voltage vector at two levels by using the reference vector to obtain an equivalent reference voltage vector; Determine a simplified control set under two levels according to the equivalent reference voltage vector, and convert the simplified control set under two levels into a vector control set under three levels; A target control vector is determined according to each control vector in the vector control set, and the three-level inverter is controlled according to the target control vector.

2. The control method of the three-level inverter according to claim 1, characterized in that: Determining a corresponding reference vector according to the positive and negative signs of the reference voltage vector in the three-phase coordinate system includes: Get symbolic function; Determine each corresponding sign function value according to the positive and negative signs corresponding to the components of each phase of the reference voltage vector in the three-phase coordinate system and the sign function; The reference vector is determined according to each corresponding sign function value.

3. The control method of the three-level inverter according to claim 1, characterized in that: Obtaining an equivalent reference voltage vector according to the reference vector and the reference voltage vector, including: The reference voltage vector is subtracted from the reference vector to obtain the equivalent reference voltage vector.

4. The control method of the three-level inverter according to claim 1, characterized in that: Determining a simplified control set under two levels according to the equivalent reference voltage vector includes: An advance vector and a lag vector of the equivalent reference voltage vector are determined, and a simplified control set under the two levels is determined according to the advance vector and the lag vector.

5. The control method of the three-level inverter according to claim 4, characterized in that: Determining a leading vector and a lagging vector of the equivalent reference voltage vector, and determining a simplified control set under the two levels according to the leading vector and the lagging vector, comprising: The action time variable is calculated based on the equivalent reference voltage vector; Determining the leading vector and lagging vector of the equivalent reference voltage vector and the action time corresponding to the action time variable according to the positive and negative signs of the action time variable; The leading vector, the lagging vector and the zero vector are determined as a simplified control set under the two levels.

6. The control method of the three-level inverter according to claim 5, characterized in that: The action time variables include a first action time variable, a second action time variable and a third action time variable; The step of calculating the action time variable based on the equivalent reference voltage vector includes: According to the equivalent reference voltage vector The projection on the axis and The projection on the axis and the DC side voltage are used to calculate the first action time variable; According to the equivalent reference voltage vector axis and the DC side voltage, and calculating the second action time variable; A Z action time variable is calculated according to the first action time variable and the second action time variable.

7. The control method of the three-level inverter according to claim 1, characterized in that: Determining a target control vector according to each control vector in the vector control set, and controlling the three-level inverter according to the target control vector, including: comparing the action time of each control vector in the vector control set, and determining the control vector with the largest action time as the target control vector; The three-level inverter is controlled according to the target control vector.

8. The control method of the three-level inverter according to any one of claims 1 to 7, characterized in that: The method further comprises: Obtaining a midpoint potential difference sampled in the current control period, and determining a polarity of the midpoint potential difference; When the polarity of the midpoint potential difference is positive, determining the redundant small vector as a negative small vector to reduce the midpoint potential difference; When the polarity of the midpoint potential difference is negative, the redundant vectorlet is determined to be a positive vectorlet to increase the midpoint potential difference.

9. A control device for a three-level inverter, characterized in that: include: Memory for storing computer programs; A processor, configured to implement the steps of the control method for a three-level inverter as claimed in any one of claims 1 to 8 when executing a computer program.

10. A three-level inverter, characterized in that: The invention comprises the control method of the three-level inverter as claimed in claim 9.

Images