Common mode voltage minimization method and system for matrix converter permanent magnet synchronous motor

Abstract

The present invention belongs to the technical field of MC-PMSM system and discloses a method for minimizing the common mode voltage of a matrix converter permanent magnet synchronous motor using model predictive control. According to the current reference value, the voltage vector required for the next cycle, i.e., the reference voltage vector V * ; According to the input voltage phase angle θ in and the phase angles of the six rotating vectors, compare the phase angle of the reference voltage vector with the phase angles of each rotating vector, and determine the sector where the reference voltage vector is located; according to the sector where the reference voltage vector is located and θ in , the obtained vectors are formed into a candidate vector set; each vector in the candidate vector set is substituted into the value function, the obtained function values ​​are compared, and the vector with the smallest function value is selected as the voltage vector of the next control cycle MC. The present invention improves the steady-state performance of torque and current while eliminating the common mode voltage.

Description

Technical Field

[0001] The invention belongs to the technical field of MC-PMSM systems, and in particular relates to a method for minimizing the common mode voltage of a matrix converter permanent magnet synchronous motor using model predictive control. Background Art

[0002] As energy issues have received more and more attention worldwide, high-performance motors and their drive technologies have become a crucial research hotspot in the energy field. Permanent magnet synchronous motors (PMSM) have the advantages of small structure, high efficiency, and high power density, and are widely used in high-end equipment manufacturing industries such as rail transportation, ship propulsion, and aerospace. The converters used to drive PMSM include two-level inverters, three-level inverters, and matrix converters (MC). Among them, the matrix converter has received widespread attention from scholars due to its advantages such as eliminating energy storage capacitors, four-quadrant operation, and low harmonic pollution.

[0003] In order to improve the performance of MC-PMSM system, scholars have applied various control algorithms to PMSM, such as direct torque control, sliding mode variable structure control, adaptive control, model predictive control (MPC), etc. Among them, direct torque control has a large flux observation error at low speed, which will lead to large torque pulsation; sliding mode variable structure control generally has a jitter problem, which affects the stability of the system; adaptive control is computationally complex and has high hardware requirements. MPC has become a new research hotspot due to its unique advantages such as simple control structure, good dynamic performance, and suitability for multi-objective control and multiple constraints. In recent years, domestic and foreign scholars have combined MPC with different theories and made many expansions in PMSM control systems.Literature MMIsmail, W.Xu, J.Ge, et.al. Adaptive Linear Predictive Model of anImproved Predictive Control of Permanent Magnet Synchronous Motor OverDifferent Speed ​​Regions[J].IEEE Transactions on Power Electronics, 2022, 37(12): 15338-15355. An adaptive MPC control algorithm is proposed to update the predictive control model in real time to reduce the ripple current; Literature S.Xu, Z.Sun, C.Yao, et.al. Model Predictive Control With Constant Switching Frequency for Three-Level T-Type Inverter-Fed PMSM Drives[J].IEEE Transactions on Industrial Electronics, 2022, 69(9): 8839-8850. In view of the problem of non-fixed switching frequency of traditional MPC, a fixed switching frequency is achieved by extending the periodic control technology; Literature Y.Zhang, Y.Huangfu, B.Tan, et.al. A Virtual MPC-Based Artificial Neural Network Controller for PMSM Drives in Aircraft Electric Propulsion System[C].2022IEEEIndustry Applications Society Annual Meeting(IAS),Detroit,MI,USA,2022.A neural network algorithm is added to the traditional MPC controller to reduce the computational burden and resource consumption by offline training of a large amount of data.

[0004] MPC has also been applied to the MC-driven PMSM system. At this time, the MPC control set includes three types of MC: effective vector, rotating vector, and zero vector. However, the effective vector and zero vector of MPC will generate a large common mode voltage (CMV). CMV will cause problems such as electromagnetic interference and leakage current, and accelerate component aging, which will have a negative impact on the permanent magnet synchronous motor itself and its drive system. If only a rotating vector is used to drive the motor, such as the MPC method proposed in L. Wang, H. Dan, Y. Zhao, et al. A Finite Control Set Model Predictive Control Method for Matrix Converter with Zero Common-mode Voltage [J]. IEEE Journal of Emerging and Selected Topics in Power Electronics, 2018, 6 (1): 327-338., CMV uses a rotating vector, which leads to large torque pulsation and current distortion.

[0005] Through the above analysis, the problems and defects of the prior art are as follows:

[0006] The application of effective vector and zero vector in traditional MPC will generate a large common-mode voltage, leading to electromagnetic interference, leakage current and other problems. If only rotating vector is used, there will be problems such as uneven distribution of rotating vector, resulting in large current harmonics and torque fluctuations. Summary of the invention

[0007] In view of the problems existing in the prior art, the present invention provides a method for minimizing the common mode voltage of a matrix converter permanent magnet synchronous motor using model predictive control.

[0008] The present invention further studies the distribution of rotating vectors, uses rotating vectors to synthesize virtual vectors to expand the candidate vector set, and synthesizes 2, 1 or 0 virtual vectors respectively under different input voltage phase conditions, so that the overall distribution of the candidate vectors tends to be uniform.

[0009] The present invention is implemented as follows: a method for minimizing the common mode voltage of a matrix converter permanent magnet synchronous motor using model predictive control, comprising the following steps:

[0010] Step 1: Calculate the voltage vector required for the next cycle based on the current reference value, i.e., the reference voltage vector V * ;

[0011] Step 2: According to the input voltage phase angle θ inand the phase angles of the six rotating vectors, compare the phase angle of the reference voltage vector with the phase angles of each rotating vector, and determine the sector where the reference voltage vector is located;

[0012] Step 3: Based on the sector where the reference voltage vector is located and θ in , the obtained vectors are combined into a candidate vector set;

[0013] Step 4: Substitute each vector in the candidate vector set into the value function, compare the obtained function values, and select the vector with the smallest function value as the voltage vector of the next control period MC.

[0014] Furthermore, the rotating vector is used to synthesize a virtual vector to expand the set of candidate vectors, and 2, 1 or 0 virtual vectors are synthesized according to the different value ranges of the angle between two adjacent rotating vectors under different input voltage phase conditions, so that the overall distribution of the candidate vectors tends to be uniform.

[0015] Furthermore, in the entire control circuit, the three-phase voltage first passes through the filter, then flows through the matrix converter, and finally is output to the permanent magnet synchronous motor; the input and output voltages of the MC satisfy the following formula:

[0016]

[0017] Among them, S ij (i∈{A, B, C}, j∈{a, b, c}) is a bidirectional switch with an output port of phase i and an input port of phase j, and S ij =1 means the switch is closed, S ij = 0 means the switch is disconnected; the switch state of MC must meet the requirements that the input terminals cannot be short-circuited and the output terminals cannot be open-circuited. Then, only one of the three switches corresponding to each phase output terminal is closed. Thus, there are 27 switch states in total, and the corresponding output voltage vector and amplitude V o , phase angle θ out and the common mode voltage V com .

[0018] Further, in step 1, record V * The component of the dq-axis rotation coordinate system is u d * ,u q * ,but:

[0019]

[0020] In the formula, i d 、i q are d and q axis currents respectively, i d * ,i q *are the corresponding current reference values ​​respectively; ω e is the rotor electrical angular velocity; R is the stator resistance; L d , L q are the equivalent inductances of d and q axes respectively; ψ m is the rotor permanent magnet flux.

[0021] Furthermore, among the six rotation vectors in step 2, +10 to +12 are positive rotations, i.e., counterclockwise rotations; -10 to -12 are reverse rotations, i.e., clockwise rotations.

[0022] Further, in step 4, the value function CF is defined as:

[0023]

[0024] Another object of the present invention is to provide a common mode voltage minimization method of a matrix converter permanent magnet synchronous motor using model predictive control and a common mode voltage minimization system of a matrix converter permanent magnet synchronous motor using model predictive control, comprising:

[0025] The voltage vector calculation module calculates the voltage vector required for the next cycle according to the current reference value, that is, the reference voltage vector V * ;

[0026] The sector determination module is based on the input voltage phase angle θ in and the phase angles of the six rotating vectors, compare the phase angle of the reference voltage vector with the phase angles of each rotating vector, and determine the sector where the reference voltage vector is located;

[0027] The candidate vector set acquisition module is based on the sector where the reference voltage vector is located and θ in , the obtained vectors are combined into a candidate vector set;

[0028] The voltage vector selection module substitutes each vector in the candidate vector set into the value function, compares the obtained function values, and selects the vector with the smallest function value as the voltage vector of the next control cycle MC.

[0029] Another object of the present invention is to provide a computer device, which includes a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor executes the steps of the method for minimizing the common-mode voltage of a matrix converter permanent magnet synchronous motor using model predictive control.

[0030] Another object of the present invention is to provide a computer-readable storage medium storing a computer program. When the computer program is executed by a processor, the processor executes the steps of the common-mode voltage minimization method of a matrix converter permanent magnet synchronous motor using model predictive control.

[0031] Another object of the present invention is to provide an information data processing terminal, which is used to implement the common-mode voltage minimization system of the matrix converter permanent magnet synchronous motor using model predictive control.

[0032] In combination with the above technical solutions and the technical problems solved, the advantages and positive effects of the technical solutions to be protected by the present invention are as follows:

[0033] First, in order to solve the problem of large current harmonics and torque fluctuations caused by uneven distribution of rotating vectors, the rotating vector is used to synthesize virtual vectors to expand the set of candidate vectors, and 2, 1 or 0 virtual vectors are synthesized under different input voltage phase conditions, so that the overall distribution of the candidate vectors tends to be uniform. It can be seen from the experiments that the steady-state performance of the MC-PMSM system under the proposed MPC method is better than that of the conventional MPC using rotating vectors, and the elimination of CMV is also achieved.

[0034] In the experiment, the effectiveness of the proposed MPC method in the MC-PMSM system was verified. Compared with the traditional MPC method, the MPC method using virtual vector synthesis performed better in steady-state performance and also achieved the elimination of CMV. This shows that the method of the present invention can effectively solve the problem caused by uneven rotating vectors and improve the performance and stability of the system.

[0035] The technical solution of the present invention successfully solves the problems of large current harmonics and torque fluctuations caused by uneven rotation vectors in the MC-PMSM system, and achieves improved system performance and stability. This provides a more reliable solution for the application of the MC-PMSM system, with significant practical value and innovation.

[0036] Second, the present invention proposes a new MPC method based on MC rotation vector. A virtual vector synthesized by a rotation vector is added to the MPC candidate vector set, and 2, 1 or 0 virtual vectors are synthesized according to different value ranges of the angle between two adjacent rotation vectors. Compared with the existing MPC method based on rotation vector, the advantages of this method are:

[0037] 1. Eliminated common mode voltage: By synthesizing virtual vectors, the common mode voltage can be effectively eliminated, the electromagnetic interference of the system is reduced, and the stability and reliability of the system are improved.

[0038] 2. Improved steady-state performance of torque and current: By adding virtual vectors, the output torque and current of the motor can be better controlled, improving the steady-state performance of the system.

[0039] 3. Improved accuracy and efficiency of the MPC method: By introducing a new synthetic rotation vector, the output of the motor can be predicted more accurately, improving the control accuracy and efficiency of the MPC method.

[0040] In summary, the new MPC method based on MC rotating vector proposed in the present invention has high technical effects and advantages, is suitable for motor control systems, can improve the performance and stability of the system, and has good practical application value.

[0041] Third, as auxiliary evidence of the inventiveness of the claims of the present invention, it is also reflected in the following important aspects:

[0042] (1) The expected benefits and commercial value of the technical solution of the present invention after transformation are as follows: The advantages of this method are that it improves the steady-state performance and efficiency of the system, reduces costs and maintenance expenses, enhances the reliability and competitiveness of the system, and has great technical application prospects and commercial value.

[0043] 1. Improve the stability and efficiency of the power system. The technical solution of the present invention can effectively eliminate the influence of the common mode voltage and improve the steady-state performance of the power system, thereby improving the stability and efficiency of the power system.

[0044] 2. Reduce equipment maintenance costs. The technical solution of the present invention can effectively reduce the failure rate and maintenance frequency of equipment in the power system, reduce equipment maintenance costs, and improve the reliability and service life of equipment.

[0045] 3. Improve the operational safety of the power system. The technical solution of the present invention can effectively improve the operational safety of the power system, reduce the occurrence of accidents and losses, and ensure the normal operation and power supply safety of the power system.

[0046] 4. Promote the progress and development of power system control technology. The technical solution of the present invention has important technical innovation significance and practical value, and can provide new ideas and methods for the progress and development of power system control technology.

[0047] 5. Huge market potential. With the continuous development and upgrading of the power system, the technical solution of the present invention has broad market prospects and potential, and can bring considerable economic benefits and commercial value to related enterprises and institutions.

[0048] (2) In the field of power system control, the impact of common-mode voltage on the steady-state performance of the system has always been a technical problem, and traditional methods are often difficult to completely solve this problem. The method proposed in the present invention successfully eliminates the impact of common-mode voltage by optimizing the control algorithm, while improving the steady-state performance of the system, filling the technical gap in this field. Through the technical solution of the present invention, the stability, efficiency and reliability of the power system can be effectively improved, and problems that cannot be solved by traditional methods can be solved. This will provide new ideas and methods for the development of the field of power system control, promote the progress and application of related technologies, and have important technical innovation significance and practical value. Therefore, the technical solution of the present invention is of great significance to the development of related fields.

[0049] (3) The technical solution of the present invention solves a technical problem that people have been eager to solve but have never been successful: The technical solution of the present invention solves one of the technical problems that people have been eager to solve but have never been successful, that is, how to improve the steady-state performance of torque and current while eliminating common-mode voltage. The traditional MPC method based on rotating vector has limited effect in eliminating common-mode voltage and cannot achieve ideal steady-state performance. The method proposed in the present invention successfully eliminates the influence of common-mode voltage by optimizing the control algorithm, while improving the steady-state performance of torque and current, fundamentally solving this technical problem. This will have a positive impact on the stability, efficiency and reliability of the power system, and bring major breakthroughs and progress to the development of related fields. Therefore, the present invention has important technical significance and practical value.

[0050] (4) The technical solution of the present invention overcomes technical bias: The technical solution of the present invention overcomes some technical biases existing in traditional technologies. The traditional MPC method based on rotating vectors has certain limitations in eliminating common-mode voltage and cannot completely solve the impact of common-mode voltage on the steady-state performance of the system. The method proposed in the present invention optimizes the control algorithm, conducts in-depth research and optimization on the common-mode voltage problem, successfully eliminates the impact of common-mode voltage, and improves the steady-state performance of torque and current. This innovative technical solution overcomes the limitations of traditional technologies, provides a new solution for the stability and efficiency of power systems, and has important technical significance and practical value. Therefore, the technical solution of the present invention overcomes technical bias to a certain extent and brings new ideas and breakthroughs to the development of related fields.

[0051] Fourth, the present invention provides a method for minimizing the common-mode voltage of a matrix converter permanent magnet synchronous motor using model predictive control. The following is a detailed analysis of the technical details, problems solved, and technological progress of the method:

[0052] 1. Step 1: Calculate the reference voltage vector

[0053] Based on the current reference value of the motor, a mathematical model or control algorithm (such as space vector pulse width modulation, direct torque control, etc.) is used to calculate the reference voltage vector `V*` required for the next control cycle.

[0054] 2. Step 2: Determine the sector where the reference voltage vector is located

[0055] The matrix converter usually uses multiple rotating vectors to synthesize the required voltage. According to the phase angle of the input voltage and the phase angle of these rotating vectors, the sector where the reference voltage vector `V*` is located is determined by comparison.

[0056] 3. Step 3: Form a candidate vector set

[0057] Once the sector where the reference voltage vector is located is determined, several vectors can be selected from the rotation vectors in the sector to form a candidate vector set. These vectors will be used for the next step of cost function evaluation.

[0058] 4. Step 4: Value Function Evaluation and Selection

[0059] Each vector in the candidate vector set is substituted into a predefined value function, which usually considers multiple factors, such as common mode voltage, torque pulsation, current ripple, etc. By comparing the values ​​of each function, the vector that minimizes the value of the value function is selected as the voltage vector of the next control cycle `MC`.

[0060] Technical issues solved:

[0061] Minimizing common-mode voltage: Common-mode voltage is an important parameter to be concerned about in motor control because it can cause problems such as electrical corrosion of motor bearings and electromagnetic interference. By optimizing the cost function and selecting the appropriate voltage vector, the common-mode voltage can be significantly reduced, improving the reliability and life of the motor operation.

[0062] Improve control accuracy: Through model predictive control, the dynamic response of the system can be predicted in real time, and the optimal control strategy can be selected based on the prediction results, thereby improving control accuracy and efficiency.

[0063] Reducing torque ripple and current ripple: By optimizing the value function, multiple performance indicators such as torque ripple and current ripple can be considered simultaneously, thereby further improving the dynamic performance of the motor while minimizing the common-mode voltage.

[0064] Technological progress achieved:

[0065] Flexibility of control strategies: Model predictive control allows trade-offs between multiple performance indicators to select the control strategy that best suits the current operating conditions.

[0066] Optimizing motor performance: By reducing common-mode voltage, torque ripple, and current ripple, the motor’s operating efficiency and reliability can be significantly improved.

[0067] Improve system robustness: Model predictive control can predict the dynamic response of the system in real time and has strong robustness to system parameter changes and external disturbances.

[0068] Expanded application scope: This method is not only applicable to permanent magnet synchronous motors, but can also be extended to other types of motors and control systems, providing a new optimization method for the field of motor control. BRIEF DESCRIPTION OF THE DRAWINGS

[0069] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the embodiments of the present invention. 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 creative work.

[0070] Figure 1 It is a flow chart of a common mode voltage minimization method of a matrix converter permanent magnet synchronous motor using model predictive control provided by an embodiment of the present invention;

[0071] Figure 2 It is a structural diagram of a common-mode voltage minimization system of a matrix converter permanent magnet synchronous motor using model predictive control provided by an embodiment of the present invention;

[0072] Figure 3 is a structural block diagram of a matrix converter provided by an embodiment of the present invention;

[0073] Figure 4 is a distribution diagram of the instantaneous position of a rotation vector provided by an embodiment of the present invention;

[0074] Figure 5 is a synthetic schematic diagram of a virtual vector provided by an embodiment of the present invention;

[0075] Figure 6 is a common mode voltage waveform diagram provided by an embodiment of the present invention;

[0076] Figure 7 It is a steady-state experimental waveform diagram provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0077] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0078] The detailed working principle of the method provided in the embodiment of the present invention involves the following steps:

[0079] Step 1: Calculate the reference voltage vector

[0080] In this step, the system calculates the voltage vector \(V^*\) required for the next control cycle based on the given current reference value. The voltage vector is generated by the current control loop, which must drive the permanent magnet synchronous motor (PMSM) appropriately in order to achieve the predetermined motor performance indicators such as torque and speed. The calculation of \(V^*\) mainly depends on the dynamic model of the motor and the current motor state (such as rotor position, speed and current).

[0081] Step 2: Determine the sector where the reference voltage vector is located

[0082] This step involves comparing the phase angle of the reference voltage vector \(V^*\) with the phase angles of the six rotating vectors output by the matrix converter to determine the sector in which \(V^*\) is located. The six rotating vectors of the matrix converter correspond to six different switching states, each of which produces a voltage vector in a different direction. By comparing the phase angle of \(V^*\) with the phase angles of these vectors, it can be determined which rotating vector \(V^*\) is closest to, thereby determining the sector in which it is located.

[0083] Step 3: Form a candidate vector set

[0084] According to the sector where \(V^*\) is located, several vectors that are most likely to achieve the required \(V^*\) are selected to form an alternative vector set. These vectors include the main vector (the dominant rotating vector in the sector) and possible zero vectors (a state where no voltage output is generated). The selection of the alternative vector set is based on the consideration of achieving precise control and reducing system losses.

[0085] Step 4: Select the best control vector

[0086] The last step is to evaluate the performance of each vector in the candidate vector set, which is usually done using a value function. This value function is usually designed based on parameters such as voltage error, power loss, and common-mode voltage. The purpose is to select an optimal control vector to minimize the common-mode voltage, thereby reducing electromagnetic interference and improving system efficiency. Each candidate vector is substituted into the value function to calculate a function value, and the vector with the smallest function value is selected as the voltage vector for the next control cycle.

[0087] Through this model predictive control (MPC)-based approach, the common-mode voltage in the permanent magnet synchronous motor system can be effectively minimized, optimizing the motor's drive efficiency and performance. This control strategy is particularly suitable for applications requiring high precision and high performance, such as electric vehicles and precision machinery control.

[0088] like Figure 1 As shown, the common mode voltage minimization method of a matrix converter permanent magnet synchronous motor using model predictive control provided by an embodiment of the present invention comprises the following steps:

[0089] Step 1: Calculate the voltage vector required for the next cycle based on the current reference value, i.e., the reference voltage vector V * ;

[0090] Step 2: According to the input voltage phase angle θ in and the phase angles of the six rotating vectors, compare the phase angle of the reference voltage vector with the phase angles of each rotating vector, and determine the sector where the reference voltage vector is located;

[0091] Step 3: Based on the sector where the reference voltage vector is located and θ in , the obtained vectors are combined into a candidate vector set;

[0092] Step 4: Substitute each vector in the candidate vector set into the value function, compare the obtained function values, and select the vector with the smallest function value as the voltage vector of the next control period MC.

[0093] Furthermore, the rotating vector is used to synthesize a virtual vector to expand the set of candidate vectors, and 2, 1 or 0 virtual vectors are synthesized according to the different value ranges of the angle between two adjacent rotating vectors under different input voltage phase conditions, so that the overall distribution of the candidate vectors tends to be uniform.

[0094] Further, in step 1, record V * The component of the dq-axis rotation coordinate system is u d * ,u q * ,but:

[0095]

[0096] In the formula, i d 、i q are d and q axis currents respectively, i d * ,i q * are the corresponding current reference values ​​respectively; ω e is the rotor electrical angular velocity; R is the stator resistance; L d , L q are the equivalent inductances of d and q axes respectively; ψ m is the rotor permanent magnet flux.

[0097] Further, in step 4, the value function CF is defined as:

[0098]

[0099] like Figure 2 As shown, an embodiment of the present invention provides a common-mode voltage minimization method for a matrix converter permanent magnet synchronous motor using model predictive control and a common-mode voltage minimization system for a matrix converter permanent magnet synchronous motor using model predictive control, comprising:

[0100] The voltage vector calculation module calculates the voltage vector required for the next cycle according to the current reference value, that is, the reference voltage vector V * ;

[0101] The sector determination module is based on the input voltage phase angle θ in and the phase angles of the six rotating vectors, compare the phase angle of the reference voltage vector with the phase angles of each rotating vector, and determine the sector where the reference voltage vector is located;

[0102] The candidate vector set acquisition module is based on the sector where the reference voltage vector is located and θ in , the obtained vectors are combined into a candidate vector set;

[0103] The voltage vector selection module substitutes each vector in the candidate vector set into the value function, compares the obtained function values, and selects the vector with the smallest function value as the voltage vector of the next control cycle MC.

[0104] 1Matrix Converter

[0105] The matrix converter is composed of 9 bidirectional switches, each of which connects any one of the three input terminals to any one of the three output terminals. Figure 3 .

[0106] In the entire control circuit, the three-phase voltage first passes through the filter, then flows through the matrix converter, and finally outputs to the permanent magnet synchronous motor. The input and output voltages of the MC satisfy the following formula:

[0107]

[0108] Among them, S ij (i∈{A, B, C}, j∈{α, b, c}) is a bidirectional switch with an output port of phase i and an input port of phase j, and S ij =1 means the switch is closed, S ij =0 means the switch is disconnected. The switch state of MC must meet the requirements that the input terminals cannot be short-circuited and the output terminals cannot be open-circuited. Then, only one of the three switches corresponding to each phase output terminal is closed. Thus, there are 27 switch states in total, and the corresponding output voltage vector and amplitude V o , phase angle θ out and the common mode voltage V comAs shown in Table 1. All voltage vectors can be divided into three categories: effective vector, rotating vector, and zero vector. Among them, ±1 to ±9 are effective vectors, their directions are fixed, but the amplitude changes with time; ±10 to ±12 are rotating vectors, their amplitude is fixed, but the phase angle changes with the input voltage angle θ in And changes.

[0109] Table 1 Matrix converter rotation vector

[0110]

[0111] Since the common mode voltage of the rotating vector is 0, the model predictive control with the rotating vector as the control set can eliminate the common mode voltage. However, due to the small number of rotating vectors and their uneven distribution, it is necessary to appropriately use the rotating vectors to synthesize virtual vectors to expand the control set.

[0112] 2 New Model Predictive Control

[0113] 2.1 Rotation vector distribution characteristics

[0114] Among the 6 rotation vectors in Table 1, +10 to +12 are positive rotations, that is, counterclockwise rotations; -10 to -12 are reverse rotations, that is, clockwise rotations. Figure 4 (a) to (e) are six rotation vectors at θ in The instantaneous position distributions are π / 12, π / 6, π / 4, π / 3 and 5π / 12 respectively. Figure 4 As can be seen, the directions of the six rotation vectors not only change with θ in The value of is constantly changing, and the distribution is also constantly changing, that is, the angle between adjacent rotation vectors is constantly changing. The rays where the 6 rotation vectors are located divide the entire plane into 6 sectors, and the boundary of each sector is two adjacent rotation vectors. The size of each sector changes with θ in The value of is constantly changing, and the rotation vectors that make up the sector boundaries are also changing, but it can be seen that the center line of each sector (i.e., the angle bisector of two adjacent rotation vectors) is fixed, located in the six directions of 0, π / 3, 2π / 3, π, 4π / 3, and 5π / 4, respectively, and are recorded as d1 to d6. For example, the sector with d1 as the center line has a boundary at θ in When it is π / 12, it is +10 and -10 two vectors, such as Figure 4 (a) and at θ in When it is 5π / 12, it is +11 and -11, two vectors, such as Figure 4 (e) as shown.

[0115] Table 2 shows θ in The rotation vector corresponding to the boundary of each sector with d1~d6 as the center line in different value intervals.

[0116] 2.2 Synthesis of Virtual Vectors

[0117] The sectors with d1 to d6 as the center line are denoted as

[0118] S1~S6. It is not difficult to see that the angle of each sector (i.e. the angle between two adjacent rotating vectors) is 0 at the minimum and 2π / 3 at the maximum, and the angles of the odd-numbered sectors (i.e. S1, S3, S5) are always equal, and the angles of the even-numbered sectors (i.e. S2, S4, S6) are always equal, while the sum of the angles of the odd-numbered sectors and the odd-numbered sectors remains unchanged at 2π / 3. The larger the angle of the sector, the sparser the vector and the larger the control error. Therefore, when the sector angle is large, the present invention adopts the method of adding virtual vectors to expand the set of candidate vectors, thereby improving the control accuracy. The principle of adding virtual vectors is: when the angle of the sector is less than 2π / 9, no virtual vector is added; when the angle of the sector is between 2π / 9 and 4π / 9, one virtual vector is added; when the angle of the sector is between 4π / 9 and 2π / 3, two virtual vectors are added, thereby making the vector distribution on the entire plane more uniform. When adding a virtual vector, two rotating vectors at the boundary of the sector are used to synthesize the virtual vector with a duty cycle of 50% each; when adding two virtual vectors, two rotating vectors are used to synthesize the two virtual vectors with a duty cycle of 1:2 and a duty cycle of 2:1 respectively. Since the sum of the angles between the odd-numbered sector and the odd-numbered sector is 2π / 3, when the angle of the odd-numbered sector is less than 2π / 9, if no virtual vector is added, the angle of the even-numbered sector must be between 4π / 9 and 2π / 3, that is, 2 virtual vectors are added; vice versa. When the angle of the odd-numbered sector is between 2π / 9 and 4π / 9, that is, when 1 virtual vector is added, the angle of the even-numbered sector must also be between 2π / 9 and 4π / 9, that is, 1 virtual vector is also added; vice versa. Therefore, the number of virtual vectors added at any time is 6.

[0119] For example, when θ in =π / 12, such as Figure 5 As shown in (a), the angle between S1, S3, and S5 is π / 6, which is less than 2π / 9, and no virtual vector is added; while the angle between S2, S4, and S6 is π / 2, which is between 4π / 9 and 2π / 3, and two virtual vectors are added. Taking the S2 sector as an example, the virtual vector is synthesized by two vectors +10 and -11. When the duty ratios of the two vectors are 2 / 3 and 1 / 3 respectively, the synthesized virtual vector is recorded as V +10 / -11 ; When the duty ratios of the two are 1 / 3 and 2 / 3 respectively, the synthesized virtual vector is recorded as V -11 / +10 According to Table 1, it can be written as a complex function:

[0120]

[0121] The other four virtual vectors can be obtained in the same way.

[0122] Table 2 Sector boundary rotation vector

[0123]

[0124] When θ in =π / 4, such as Figure 5 As shown in (c), the situation is just the opposite. The angle between S2, S4, and S6 is π / 6, and no virtual vector is added; while the angle between S1, S3, and S5 is π / 2, and 2 virtual vectors are added. The naming rules of virtual vectors are similar.

[0125] When θ in =π / 6, such as Figure 5 As shown in (b), the angles of the six sectors are all π / 3, ranging from 2π / 9 to 4π / 9, and one virtual vector is added to each sector. Taking sector S2 as an example, the virtual vector is synthesized by two vectors +10 and -11, and the duty ratio of the two vectors is 1 / 2. The synthesized virtual vector is recorded as V +10，-11 According to Table 1, it can be written as a complex function:

[0126]

[0127] Therefore, for any value of θ in The number of virtual vectors and the synthesis method of each sector can be determined. Table 3 shows θ in The rotation vectors of S1~S6 and the synthesized virtual vector within different value ranges.

[0128] Table 3 Optional vectors for each sector

[0129]

[0130]

[0131] In order to prove the creativity and technical value of the technical solution of the present invention, application verification can be carried out on the following specific products or related technologies:

[0132] In the electric vehicle drive system, the virtual vector synthesis method of the present invention can be applied to the permanent magnet synchronous motor (PMSM) drive system of the electric vehicle. The method can effectively reduce current harmonics and torque fluctuations, improve the efficiency and stability of the system, and extend the cruising range of the electric vehicle.

[0133] The MPC method of the present invention is used in an industrial robot control system to optimize the robot's motion control, reduce vibration, and improve working accuracy and production efficiency.

[0134] In a wind power generation system, the virtual vector synthesis technology of the present invention can reduce the harmonic distortion of the system, improve power generation efficiency and stability, and reduce maintenance costs.

[0135] By using the MPC method of the present invention in an industrial variable frequency air conditioning system, the control strategy of the system can be optimized, energy consumption and operating costs can be reduced, and the performance and reliability of the system can be improved.

[0136] Through the application embodiments on the above specific products or related technologies, it can be further proved that the technical solution of the present invention has creativity and technical value, and has broad application prospects and market potential.

[0137] The embodiments of the present invention have achieved some positive effects during the development or use process, and indeed have great advantages over the prior art. The following content is described in conjunction with data, charts, etc. of the test process.

[0138] In order to verify the effectiveness and feasibility of the proposed MPC method, an experiment was conducted using the MC-PMSM system. The method in the literature L. Wang, H. Dan, Y. Zhao, et al. A Finite Control Set Model Predictive Control Method for Matrix Converter with Zero Common-mode Voltage [J]. IEEE Journal of Emerging and Selected Topics in Power Electronics, 2018, 6 (1): 327-338. is referred to as traditional MPC for comparison. The relevant parameters of PMSM are shown in Table 4. The control circuit of MC consists of TMS320F28335 DSP and EP4CE6 FPGA. The program of the motor control algorithm runs in the DSP, while the commutation program of MC runs in the FPGA.

[0139] Table 4 Permanent magnet motor parameters

[0140]

[0141] Figure 6 (a) shows the common-mode voltage waveform of the traditional MPC. Figure 6 (b) shows the common mode voltage waveform of the MPC of the present invention. Since both methods use rotating vector drive motors, the common mode voltage is effectively suppressed.

[0142] Figure 6 (a) and 6(b) show the steady-state experimental waveforms of the conventional MPC and the MPC of the present invention at 400 r / min, respectively, including torque, MC output current and input current.

[0143] Comparison between the two shows that in terms of steady-state control, the MPC proposed in the present invention has a smoother motor torque and a more sinusoidal current waveform than the traditional MPC. This is because the method of the present invention uses a virtual vector synthesized by a rotating vector, which effectively improves the steady-state performance of the motor.

[0144] in conclusion

[0145] The present invention proposes a novel MPC using a rotating vector and its synthesized virtual vector to control the MC-PMSM system. Through theoretical analysis and experimental verification, it is proved that the MPC proposed in the present invention can eliminate CMV, and compared with the traditional rotating vector MPC, the MPC of the present invention significantly suppresses torque fluctuations and current harmonics, and improves the steady-state performance of the motor system.

[0146] The above description is only a specific implementation mode of the present invention, but the protection scope of the present invention is not limited thereto. Any modifications, equivalent substitutions and improvements made by any technician familiar with the technical field within the technical scope disclosed by the present invention and within the spirit and principle of the present invention should be covered by the protection scope of the present invention.

Claims

1. A method for minimizing the common mode voltage of a matrix converter permanent magnet synchronous motor using model predictive control, characterized in that: The following steps are involved: Step 1: Calculate the voltage vector required for the next cycle based on the current reference value, i.e., the reference voltage vector V * ; Note V * The component of the dq-axis rotation coordinate system is u d * ,u q * ,but: In the formula, i d 、i q are the d and q axis currents respectively, i d * ,i q * are the corresponding current reference values ​​respectively; ω e is the rotor electrical angular velocity; R is the stator resistance; L d , L q are the equivalent inductances of d and q axes respectively; ψ m is the rotor permanent magnet flux; Step 2: According to the input voltage phase angle θ in and the phase angles of the six rotating vectors, compare the phase angle of the reference voltage vector with the phase angles of each rotating vector, and determine the sector where the reference voltage vector is located; among the six rotating vectors, +10 to +12 are positive rotations, i.e., counterclockwise rotation; -10 to -12 are reverse rotations, i.e., clockwise rotation; Step 3: Based on the sector where the reference voltage vector is located and θ in , the obtained vectors are combined into a candidate vector set; the candidate vector set is expanded by synthesizing virtual vectors using rotating vectors, and 2, 1 or 0 virtual vectors are synthesized respectively according to different value ranges of the angle between two adjacent rotating vectors under different input voltage phase conditions, so that the overall distribution of the candidate vectors tends to be uniform; the principle of adding virtual vectors is: when the angle of the sector is less than 2π / 9, no virtual vector is added; when the angle of the sector is between 2π / 9 and 4π / 9, one virtual vector is added; when the angle of the sector is between 4π / 9 and 2π / 3, two virtual vectors are added, thereby making the vector distribution on the entire plane more uniform; when adding a virtual vector, the two rotating vectors at the boundary of the sector are used to synthesize the virtual vector with a duty cycle of 50% each; when adding two virtual vectors, the two rotating vectors are used to synthesize two virtual vectors with a duty cycle of 1:2 and a duty cycle of 2:1 respectively; the number of virtual vectors added at any time is 6; Step 4: Substitute each vector in the candidate vector set into the cost function, compare the obtained function values, and select the vector with the smallest function value as the voltage vector of the next control cycle MC; the cost function CF is defined as:

2. The method for minimizing the common mode voltage of a matrix converter permanent magnet synchronous motor using model predictive control according to claim 1, characterized in that: In the entire control circuit, the three-phase voltage first passes through the filter, then flows through the matrix converter, and finally is output to the permanent magnet synchronous motor; the input and output voltages of the MC satisfy the following formula: Among them, S ij is a bidirectional switch with an output port of phase i and an input port of phase j, where i∈{A, B, C}, j∈{a, b, c}, and S ij =1 means the switch is closed, S ij =0 means the switch is open; there is only one switch closed in the three switches corresponding to each phase output terminal, and there are 27 switch states in total. The corresponding output voltage vector and amplitude V o , phase angle θ out and the common mode voltage V com .

3. A common mode voltage minimization system for a matrix converter permanent magnet synchronous motor using model predictive control according to any one of claims 1 to 2, characterized in that: include: The voltage vector calculation module calculates the voltage vector required for the next cycle according to the current reference value, that is, the reference voltage vector V * ; The sector determination module is based on the input voltage phase angle θ in and the phase angles of the six rotating vectors, compare the phase angle of the reference voltage vector with the phase angles of each rotating vector, and determine the sector where the reference voltage vector is located; The candidate vector set acquisition module is based on the sector where the reference voltage vector is located and θ in , the obtained vectors are combined into a candidate vector set; The voltage vector selection module substitutes each vector in the candidate vector set into the value function, compares the obtained function values, and selects the vector with the smallest function value as the voltage vector of the next control cycle MC.

4. A computer device, characterized in that: The computer device includes a memory and a processor, the memory stores a computer program, and when the computer program is executed by the processor, the processor executes the steps of the common-mode voltage minimization method of a matrix converter permanent magnet synchronous motor using model predictive control as described in any one of claims 1 to 2.

5. A computer-readable storage medium, characterized in that: A computer program is stored, and when the computer program is executed by a processor, the processor executes the steps of the method for minimizing the common-mode voltage of a matrix converter permanent magnet synchronous motor using model predictive control as described in any one of claims 1 to 2.

6. An information data processing terminal, characterized in that: The information data processing terminal is used to implement the common mode voltage minimization system of the matrix converter permanent magnet synchronous motor using model predictive control as described in claim 3.

Images