Low current harmonic direct torque control method for dual three-phase permanent magnet synchronous motor

Abstract

This invention relates to a low-current harmonic direct torque control method for a dual three-phase permanent magnet synchronous motor (PMSM), belonging to the field of PMSM control technology. The method is as follows: S1: Establish mathematical models of the fundamental and harmonic subspaces of the dual three-phase PMSM under vector space decoupling, and clarify the voltage vector distribution characteristics of the two subspaces; S2: In the fundamental subspace, design a duty cycle modulation strategy based on the principle of minimizing the error vector, and an improved deadbeat flux linkage torque controller; S3: In the harmonic subspace, design an online learning harmonic canceller, using the least mean square algorithm to update the weight coefficients in real time; S4: Select the D1 vector set as the effective vector of the harmonic subspace and perform sector division; S5: Add the effective voltage vector action time of the fundamental and harmonic subspaces to generate the final switching sequence, which is output to the dual three-phase inverter to effectively suppress torque ripple and harmonic current of the dual three-phase PMSM.

Description

Technical Field

[0001] This invention belongs to the field of permanent magnet synchronous motor control technology, and in particular, it is a method for low current harmonic direct torque control of a dual three-phase permanent magnet synchronous motor. Background Technology

[0002] Dual three-phase permanent magnet synchronous motors (PMSMs) have been widely used in electrical drive fields with high power density and high reliability requirements, such as aerospace, ship propulsion, and electric vehicles, due to their advantages of low voltage, high power, high reliability, and strong fault tolerance. Currently, vector space decoupling-based modeling and control is the mainstream method for achieving high-performance operation of dual three-phase PMSMs. This method decouples the variables of the dual three-phase PMSMs with isolated neutral points to mutually independent fundamental and harmonic subspaces.

[0003] Direct torque control (DTC), a simple and fast-responding motor control strategy, has been introduced into dual three-phase permanent magnet synchronous motor drive systems. Traditional DTC strategies directly generate switching signals based on flux linkage and torque hysteresis controllers. Compared to the rotating coordinate transformation and cascaded current regulator design required in field-oriented control (DOC), its control structure is simpler, its dynamic response is faster, and it exhibits a certain degree of robustness to changes in motor parameters. However, when applying this strategy to dual three-phase permanent magnet synchronous motors, coordinate transformation based on vector space decoupling methods is required to separate the fundamental and harmonic subspaces. Furthermore, traditional DTC lacks an active current regulation mechanism in the harmonic subspace, making it difficult to effectively suppress fifth and seventh harmonic currents. These harmonic currents, present as AC components in the harmonic subspace, exacerbate torque ripple and system losses.

[0004] To improve the performance of dual three-phase permanent magnet synchronous motor systems based on direct torque control, existing technologies mainly focus on two aspects. One approach focuses on optimizing the fundamental subspace, such as eliminating the influence of the harmonic subspace by constructing a virtual voltage vector or introducing duty cycle modulation strategies to reduce torque ripple. However, these methods often fail to adequately consider the active suppression of harmonic subspace currents, resulting in significant harmonic currents. Another approach attempts to introduce a resonant controller in the harmonic subspace to suppress AC harmonics, but resonant controllers have many parameters, complex adjustment processes, and their performance is highly dependent on the accuracy of the resonant frequency, exhibiting poor adaptability under dynamic conditions such as speed variations. Furthermore, most existing improved direct torque control schemes utilize hysteresis controllers, whose inherent nonlinearity and variable switching frequency limit further improvements in control accuracy.

[0005] Therefore, how to effectively suppress torque ripple and harmonic current of dual three-phase permanent magnet synchronous motors while maintaining the fast dynamic response of direct torque control, and simplify the controller structure, is a key problem that urgently needs to be solved in this technical field. Summary of the Invention

[0006] The purpose of this invention is to solve the problems of large torque ripple and difficulty in suppressing harmonic current in the traditional direct torque control in the dual three-phase permanent magnet synchronous motor drive system. At the same time, it improves the limitations of existing improved direct torque control with many tuning parameters and poor dynamic adaptability when introducing a resonant controller, and fully takes into account the coordinated control requirements of the fundamental subspace and the harmonic subspace. Therefore, a low-current harmonic direct torque control method for dual three-phase permanent magnet synchronous motors is proposed.

[0007] This invention presents an enhanced direct torque control scheme based on a combination of duty cycle modulation with error vector minimization and an online learning harmonic canceller, which is of great significance for the practical application of direct torque control strategies for dual three-phase permanent magnet synchronous motors in the field of high-performance electric drives.

[0008] 1. Establish mathematical models of the fundamental and harmonic subspaces of a dual three-phase permanent magnet synchronous motor under vector space decoupling, clarify the voltage vector distribution characteristics of the two subspaces, and lay the foundation for zoned independent control;

[0009] 2. In the fundamental subspace, this invention proposes a duty cycle modulation strategy based on the principle of minimizing the error vector, calculates the optimal action time of a single effective voltage vector (calculate the optimal action time based on formula (4), realizes accurate tracking of the reference voltage vector, and combines an improved deadbeat flux torque controller to replace the traditional hysteresis controller to reduce torque ripple;

[0010] 3. In the harmonic subspace, this invention proposes an online learning type harmonic canceller, which uses the least mean square algorithm to update the weight coefficients in real time and generate an electrical signal that can compensate for harmonics (expressed by formula (15)) to achieve effective suppression of the 5th and 7th harmonic currents;

[0011] 4. Based on the amplitude characteristics of the harmonic subspace voltage vector, this invention proposes to select four non-zero voltage vectors in the D1 vector set as effective voltage vectors, and to divide the sector based on the reference voltage phase to determine the action time of each vector (expressed by formulas (16)-(20)) to achieve coordinated modulation of harmonic current suppression and fundamental subspace control.

[0012] 5. The effective voltage vector of the fundamental subspace is added together with the effective voltage vector of the harmonic subspace to generate the final switching sequence, which is then output to the dual three-phase inverter to effectively suppress torque pulsation and harmonic current of the dual three-phase permanent magnet synchronous motor.

[0013] The low current harmonics refer to the reduction of phase current harmonics by adding a current harmonic suppression module (online learning type harmonic canceller) to the direct torque control strategy.

[0014] To achieve the above objectives, the technical solution adopted by the present invention is as follows:

[0015] A method for low-current harmonic direct torque control of a dual three-phase permanent magnet synchronous motor, the method comprising the following steps:

[0016] S1: Establish mathematical models of the fundamental and harmonic subspaces of the dual three-phase permanent magnet synchronous motor under vector space decoupling, and clarify the voltage vector distribution characteristics of the two subspaces, laying the foundation for zoned independent control.

[0017] S2: In the fundamental subspace, design a duty cycle modulation strategy based on the principle of minimizing the error vector, and design an improved deadbeat flux torque controller.

[0018] S3: In the harmonic subspace, design an online learning type harmonic canceller and use the least mean square algorithm to update the weight coefficients in real time.

[0019] S4: Select the D1 vector set as the effective vector of the harmonic subspace, and divide the sector based on the reference voltage phase to achieve coordinated modulation of harmonic current suppression and fundamental subspace control;

[0020] S5: The effective voltage vector of the fundamental subspace is added together with the effective voltage vector of the harmonic subspace to generate the final switching sequence, which is then output to the dual three-phase inverter to effectively suppress torque pulsation and harmonic current of the dual three-phase permanent magnet synchronous motor.

[0021] Furthermore, step S1 includes the following steps:

[0022] S11: Establish the mathematical model of the fundamental subspace and harmonic subspace of the dual three-phase permanent magnet synchronous motor under vector space decoupling, as shown in equation (1) below:

[0023]

[0024] in, , , and They are respectively axis, axis, shaft and Voltage on the shaft; , , and They are respectively axis, axis, shaft and The current in the shaft; This refers to the stator winding resistance. and They are respectively axis The inductance of the shaft; For stator winding leakage inductance; The flux linkage is along the d-axis. for Axial magnetic flux; Electric angular velocity;

[0025] S12: Clarify the voltage vector distribution characteristics of the fundamental and harmonic subspaces.

[0026] S121: A two-level dual three-phase inverter with a common DC bus outputs 64 basic voltage vectors, including 60 non-zero voltage vectors and 4 zero vectors; the 60 non-zero voltage vectors are then used to determine their values ​​based on... subplane and The projections on the subplane are grouped to obtain group D3 and group D4; The subplane is the stationary coordinate system of the fundamental subspace. The subplane is a harmonic subspace;

[0027] S122: Using a reconstructed set of virtual voltage vectors, assuming the virtual voltage vectors... The duration of action is Let the duration of action of the voltage vectors in groups D3 and D4 be respectively and The relationship is expressed as:

[0028]

[0029] in, This is the bus voltage.

[0030] Furthermore, step S2 includes the following steps:

[0031] S21: In the fundamental subspace, design a duty cycle modulation strategy based on the principle of minimizing the error vector.

[0032] set up (OA) is a given reference voltage vector with a phase angle of . According to the phase angle Determine the sector where the reference voltage vector is located; select the effective voltage vector corresponding to the sector. (OB), the effective voltage vector and The included angle of the axis is Then the duty cycle Represented as:

[0033]

[0034] in, , , ;

[0035] , They are respectively shaft and Shaft voltage;

[0036] Based on the obtained duty cycle The duration of action of the effective voltage vector in the fundamental subspace is obtained:

[0037]

[0038] in, It is an electrical cycle;

[0039] S22: Improved design of deadbeat flux torque controller;

[0040] The axial flux linkage is represented as:

[0041]

[0042] in, It is a permanent magnet flux linkage;

[0043]

[0044] Substituting equation (1) into equation (6) yields the deadbeat flux torque controller:

[0045]

[0046] For surface-mount permanent magnet synchronous motors = , and The discrete domain expression is shown below:

[0047]

[0048] in, , , Given a reference torque, To calculate the torque, Represents the current moment;

[0049] and The discrete domain expression is shown below:

[0050]

[0051] The error vector transformation duty cycle modulation required by the inverse Park transform is obtained. and .

[0052] Furthermore, step S3 includes the following steps:

[0053] S31: Design an online learning harmonic canceller in the harmonic subspace.

[0054] The 5th and 7th harmonic currents in the harmonic subspace are mapped to... In the subspace, when transformed to a reference coordinate system rotating at the fundamental frequency, a 6th harmonic current appears; based on the known 6th harmonic current, assume... and As shown below:

[0055]

[0056] in, It is an electrical cycle; , These are the cosine and sine harmonic basis functions, respectively;

[0057] The error between the harmonic signal and the compensation signal over one electrical cycle is given by the following formula:

[0058]

[0059] in, It is a 6th harmonic current. and They are respectively and Weighting coefficients;

[0060] S32: Update weight coefficients in real time using the least mean square algorithm;

[0061] The learning and updating process of the weight coefficients uses the least mean square algorithm, assuming the objective function is... Its expression is:

[0062]

[0063] The weight coefficients are obtained by minimizing the objective function using gradient descent. The gradient of the weight coefficients is expressed as:

[0064]

[0065] in, express The gradient at the current moment;

[0066] Introducing adaptive learning rate The adaptive calculation process for the weighting coefficients is as follows:

[0067]

[0068] The output signal of the online learning harmonic canceller is expressed as:

[0069] .

[0070] Furthermore, in step S4, the selection of the D1 vector set as the effective vector of the harmonic subspace, and the sector division based on the reference voltage phase, realizes the coordinated modulation of harmonic current suppression and fundamental subspace control; the specific steps are as follows:

[0071] Selecting the D1 vector set , , , The four voltage vectors serve as the effective voltage vectors. When the harmonic subspace reference voltage is located in sector I, , The durations of the voltage vectors are as follows:

[0072]

[0073]

[0074] in, and They are respectively shaft and Shaft voltage;

[0075] When the harmonic subspace reference voltage is located in sector III , The durations of action of the voltage vectors are as follows:

[0076]

[0077]

[0078] Furthermore, in step S5, the effective voltage vector action time of the fundamental subspace is added to the effective voltage vector action time of the harmonic subspace to generate the final switching sequence, which is then output to the dual three-phase inverter to effectively suppress torque ripple and harmonic current of the dual three-phase permanent magnet synchronous motor; the specific steps are as follows:

[0079] According to equations (4), (16), and (17), the effective voltage vector duration within one electrical cycle is:

[0080]

[0081] Alternatively, according to equations (4), (18), and (19), the effective voltage vector's duration within one electrical cycle can be obtained as follows:

[0082]

[0083] Based on the above, a switching sequence is generated to effectively suppress torque ripple and harmonic current in a dual three-phase permanent magnet synchronous motor.

[0084] The advantages of this invention over the prior art are:

[0085] 1. The advantage of this invention is that it provides a direct torque control scheme for a dual three-phase permanent magnet synchronous motor based on an error vector minimization duty cycle regulator and an online learning wave canceller, which has better suppression performance for torque ripple and harmonic current.

[0086] 2. The key feature of this invention is the implementation of an enhanced direct torque control scheme, simultaneously optimizing control in both the fundamental and harmonic subspaces. In the fundamental subspace, torque ripple is reduced through an improved deadbeat flux linkage torque controller and a proposed error vector minimization duty cycle modulation. In the harmonic subspace, periodic harmonic currents are suppressed through a proposed online learning harmonic canceller. Therefore, both the steady-state accuracy and dynamic response of the dual three-phase permanent magnet synchronous motor system are improved. The collaborative control scheme balances the independence of the two subspaces, and torque ripple and harmonic suppression are achieved using a single control framework, resulting in good overall control performance.

[0087] 3. This scheme is simple to implement. It only requires adding error vector minimization duty cycle calculation and online learning harmonic canceller to the traditional direct torque control, and integrating the calculation results of the two. This can enhance the suppression performance of direct torque control strategy in torque ripple and current harmonics. Compared with the traditional direct torque control method that does not consider active suppression of harmonic subspace, it can significantly improve the steady-state operation quality and dynamic response capability of the motor, while maintaining the inherent advantages of direct torque control structure and fast dynamic response. Attached Figure Description

[0088] Figure 1 This is the overall control block diagram of the present invention.

[0089] Figure 2 For non-zero vector projections, where:

[0090] Figure 2 (a) is Non-zero vector projection diagram of the subplane;

[0091] Figure 2 (b) is Non-zero vector projection diagram of the subplane.

[0092] Figure 3 To reconstruct the virtual vector projection map, where:

[0093] Figure 3 (a) is Subplane reconstruction of virtual vector projection map;

[0094] Figure 3 (b) is Subplane reconstruction of virtual vector projection map.

[0095] Figure 4 This is a schematic diagram of the duty cycle modulation principle for minimizing the error vector.

[0096] Figure 5 This is a schematic diagram of an online learning type harmonic canceller.

[0097] Figure 6 for Schematic diagram of subspace sector division and voltage vector action time determination;

[0098] Figure 7 This is a flowchart for low-current harmonic direct torque control.

[0099] Figure 8 The graph shows the steady-state performance results of the conventional method and the present invention, wherein:

[0100] Figure 8 (a) shows the steady-state performance results of the traditional method;

[0101] Figure 8 (b) is a graph showing the steady-state performance results of the present invention.

[0102] Figure 9 The graph shows the FFT analysis results of the A-phase motor using the traditional method and the present invention. THD in the graph represents total harmonic distortion.

[0103] Figure 9 (a) is a graph showing the FFT analysis results of the A-phase motor using the traditional method;

[0104] Figure 9(b) is a graph showing the FFT analysis results of the A-phase motor of the present invention.

[0105] Figure 10 The diagram shows the dynamic performance results of the existing resonant direct torque control method and the embodiment of the present invention under varying loads, wherein:

[0106] Figure 10 (a) is a dynamic result diagram of the existing resonant direct torque control method;

[0107] Figure 10 (b) is a diagram showing the dynamic performance results of the variable load in an embodiment of the present invention. Detailed Implementation

[0108] Figure 1 Description: The proposed overall control block diagram is an integrated control structure for speed, current, and torque. Speed ​​and current measurement information is used for torque, flux linkage control, and harmonic current suppression. Based on the estimated torque and flux linkage, the duty cycle of the effective voltage vector is calculated using the principle of minimizing the error vector. Simultaneously, an online learning harmonic canceller module is applied to compensate for harmonic currents, providing a reference voltage, and calculating the harmonic current duty cycle using voltage vectors from four harmonic planes. The vector action time obtained from torque and flux linkage control is combined with the vector action time obtained from harmonic plane current control to obtain the final effective voltage vector action time, which is output to the inverter to ensure low harmonic current and high dynamic performance of the system.

[0109] Figure 7 Explanation: This invention decouples motor control into two independent subspaces for parallel control—the fundamental subspace is responsible for torque and flux linkage, and the harmonic subspace is responsible for suppressing harmonic current. After the effective voltage vectors of the two subspaces have been obtained for their respective durations, the results of the two subspaces are added together to finally generate a switching sequence.

[0110] Example:

[0111] This embodiment describes a method for low-current harmonic direct torque control of a dual three-phase permanent magnet synchronous motor, the method comprising the following steps:

[0112] S1: Establish mathematical models of the fundamental and harmonic subspaces of a dual three-phase permanent magnet synchronous motor under vector space decoupling, and clarify the voltage vector distribution characteristics of the two subspaces and the application of the virtual voltage vector; the specific steps are as follows:

[0113] S11: Establishing the fundamental subspace of a dual three-phase permanent magnet synchronous motor under vector space decoupling ( (within the reference frame) Harmonic subspace ( The mathematical model (within the reference frame) is shown in equation (1) below:

[0114]

[0115] in, , , and They are respectively axis, axis, shaft and Voltage on the shaft; , , and They are respectively axis, axis, shaft and The current in the shaft; This refers to the stator winding resistance. and They are respectively axis The inductance of the shaft; For stator winding leakage inductance; The flux linkage is along the d-axis. for Axial magnetic flux; Electric angular velocity;

[0116] S12: Clarify the voltage vector distribution in the fundamental and harmonic subspaces and the application of virtual voltage vectors.

[0117] S121: Two-level dual three-phase inverter with common DC bus output 64 (2 6 = 64) basic voltage vectors, which can be derived from = ( The binary value of is used to define it, where: ( = , , , , , The 64 basic voltage vectors represent the switching states of the upper arm of the inverter, including 60 non-zero voltage vectors and 4 zero vectors; the 60 non-zero voltage vectors are then used to determine their switching states based on their position on the upper arm of the inverter. subplane and The projections on the subplane are grouped to obtain group D3 and group D4. Figure 2 This demonstrates the 60 non-zero voltage vectors in subplane and in (projection on the subplane); the The subplane is the stationary coordinate system of the fundamental subspace. The subplane is a harmonic subspace;

[0118] S122: Using a reconstructed set of virtual voltage vectors, assuming the virtual voltage vectors... The duration of action is Let the duration of action of the voltage vectors in groups D3 and D4 be respectively and The relationship is expressed as:

[0119]

[0120] Among them, V dc Bus voltage ( Figure 3 The reconstructed virtual voltage vector set is shown. ).

[0121] S2: In the fundamental subspace, a duty cycle modulation strategy based on the principle of minimizing the error vector is designed, and an improved deadbeat flux torque controller is designed; the specific steps are as follows:

[0122] S21: In the fundamental subspace, design a duty cycle modulation strategy based on the principle of minimizing the error vector.

[0123] Figure 4 This diagram illustrates the principle of duty cycle modulation with error vector minimization. (Settings) (OA) is a given reference voltage vector with a phase angle of . According to the phase angle Determine the sector where the reference voltage vector is located (assuming the reference voltage vector is located in sector II, see [reference]). Figure 4 ); Select the effective voltage vector corresponding to the sector. (OB), the effective voltage vector and The included angle of the axis is Then the duty cycle Represented as:

[0124]

[0125] in, , , ;

[0126] , They are respectively shaft and Shaft voltage;

[0127] Based on the obtained duty cycle The duration of action of the effective voltage vector in the fundamental subspace is obtained:

[0128]

[0129] in, It is an electrical cycle;

[0130] S22: Improved design of deadbeat flux torque controller;

[0131] The axial flux linkage is represented as:

[0132]

[0133] in, It is a permanent magnet flux linkage;

[0134]

[0135] Substituting equation (1) into equation (6) yields the deadbeat flux torque controller:

[0136]

[0137] For surface-mount permanent magnet synchronous motors = , and The discrete domain expression is shown below:

[0138]

[0139] in, , , Given a reference torque, To calculate the torque, Represents the current moment;

[0140] and The discrete domain expression is shown below:

[0141]

[0142] The error vector transformation duty cycle modulation required can be obtained through the inverse Park transform. and .

[0143] S3: In the harmonic subspace, design an online learning harmonic canceller and use the least mean square algorithm to update the weight coefficients in real time; the specific steps are as follows:

[0144] S31: Design an online learning harmonic canceller in the harmonic subspace.

[0145] The 5th and 7th harmonic currents in the harmonic subspace are mapped to... In the subspace, when transformed to a reference coordinate system rotating at the fundamental frequency, a 6th harmonic current will appear. Figure 5 A schematic diagram of an online learning type harmonic canceller is shown. Based on the known 6th harmonic current, it is assumed that... and As shown below:

[0146]

[0147] in, It is an electrical cycle; , These are the cosine and sine harmonic basis functions, respectively;

[0148] The error between the harmonic signal and the compensation signal over one electrical cycle is given by the following formula:

[0149]

[0150] in, It is a 6th harmonic current. and They are respectively and Weighting coefficients;

[0151] S32: Update weight coefficients in real time using the least mean square algorithm;

[0152] The learning and updating process of the weight coefficients uses the least mean square algorithm, assuming the objective function is... Its expression is:

[0153]

[0154] The weight coefficients are obtained by minimizing the objective function using gradient descent. The gradient of the weight coefficients is expressed as:

[0155]

[0156] in, express The gradient at the current moment;

[0157] Introducing adaptive learning rate The adaptive calculation process for the weighting coefficients is as follows:

[0158]

[0159] The output signal of the online learning harmonic canceller is expressed as:

[0160] .

[0161] S4: Select the D1 vector set as the harmonic subspace ( The effective vector of the subspace (which ensures maximum voltage utilization during harmonic suppression) is determined, and sectors are divided based on the reference voltage phase to achieve coordinated modulation of harmonic current suppression and fundamental subspace control; the specific steps are as follows:

[0162] Selecting the D1 vector set , , , Four voltage vectors are used as effective voltage vectors (harmonic subspace sector division as follows) Figure 6 As shown), when the harmonic subspace reference voltage is located in sector I, , The durations of the voltage vectors are as follows:

[0163]

[0164]

[0165] in, and They are respectively shaft and Shaft voltage;

[0166] When the harmonic subspace reference voltage is located in sector III , The durations of action of the voltage vectors are as follows:

[0167]

[0168]

[0169] When the harmonic subspace reference voltage is located in other sectors, the corresponding effective voltage vector and its duration are shown in Table 1.

[0170] Figure 6 Showing The process of sector partitioning of the subspace, selection of effective voltage, and determination of the action time. Selection of V 14 V 34 V 49 and V 29 The four voltage vectors are used as effective voltage vectors.

[0171] Table 1. Selection of Harmonic Subspace Vector and Determination of Action Time

[0172]

[0173] S5: The effective voltage vector action time of the fundamental subspace is added to the effective voltage vector action time of the harmonic subspace to generate the final switching sequence, which is then output to the dual three-phase inverter to effectively suppress torque ripple and harmonic current of the dual three-phase permanent magnet synchronous motor; the specific steps are as follows:

[0174] According to equations (4), (16), and (17), the effective voltage vector duration within one electrical cycle is:

[0175]

[0176] Alternatively, according to equations (4), (18), and (19), the effective voltage vector's duration within one electrical cycle can be obtained as follows:

[0177]

[0178] Based on the above, a switching sequence is generated to effectively suppress torque ripple and harmonic current in a dual three-phase permanent magnet synchronous motor.

[0179] The experimental results in this embodiment are compared as follows (Table 2 shows the motor parameters used in the experiment of this embodiment):

[0180] Table 2 Motor parameters used in the embodiments

[0181]

[0182] (1) Comparison results of steady-state experiments

[0183] Under rated speed and torque conditions, the results of the traditional direct torque control method without harmonic suppression are as follows: Figure 8 (Including electromagnetic torque, A-phase and X-phase currents and) (axis harmonics) and Figure 9As shown, traditional methods exhibit significant torque ripple and harmonic current in steady-state conditions. According to the FFT analysis of the A-phase current, its total harmonic distortion (THD) is 28.83%. However, the embodiment of this invention, due to improvements in flux linkage and torque control, achieves the lowest possible torque ripple. Simultaneously, the online learning harmonic canceller effectively suppresses harmonic current, resulting in a THD of 15.24%, a reduction of 13.59%.

[0184] (2) Comparison results of dynamic tests under variable load

[0185] Under variable load dynamic conditions, a comparative experiment was conducted between the resonant direct torque control method and the method described in this invention, see [link to experiment]. Figure 10 (Including electromagnetic torque, (Shaft current and flux linkage). When a dual three-phase permanent magnet synchronous motor is running at rated speed, a sudden application of rated torque is performed, followed by load removal after 2 seconds. Both methods exhibit stable harmonic current suppression performance during load changes, but the resonant controller requires tuning more than two unknown parameters, while this invention only requires tuning one learning parameter. Furthermore, due to the error vector minimization duty cycle modulation strategy of this embodiment, the torque ripple of this embodiment is reduced.

Claims

1. A method for low current harmonic direct torque control of a dual three-phase permanent magnet synchronous motor, characterized by: The method includes the following steps: S1: Establish mathematical models of the fundamental and harmonic subspaces of the dual three-phase permanent magnet synchronous motor under vector space decoupling, and clarify the voltage vector distribution characteristics of the two subspaces, laying the foundation for zoned independent control. S2: In the fundamental subspace, design a duty cycle modulation strategy based on the principle of minimizing the error vector, and design an improved deadbeat flux torque controller. S3: In the harmonic subspace, design an online learning type harmonic canceller and use the least mean square algorithm to update the weight coefficients in real time. S4: Select the D1 vector set as the effective vector of the harmonic subspace, and divide the sector based on the reference voltage phase to achieve coordinated modulation of harmonic current suppression and fundamental subspace control; S5: The effective voltage vector of the fundamental subspace is added together with the effective voltage vector of the harmonic subspace to generate the final switching sequence, which is then output to the dual three-phase inverter to effectively suppress torque pulsation and harmonic current of the dual three-phase permanent magnet synchronous motor.

2. The control method according to claim 1, characterized by: Step S1 includes the following steps: S11: Establish the mathematical model of the fundamental subspace and harmonic subspace of the dual three-phase permanent magnet synchronous motor under vector space decoupling, as shown in equation (1) below: in, , , and They are respectively axis, axis, shaft and Voltage on the shaft; , , and They are respectively axis, axis, shaft and The current in the shaft; This refers to the stator winding resistance. and They are respectively axis The inductance of the shaft; For stator winding leakage inductance; The flux linkage is along the d-axis. for Axial magnetic flux; Electric angular velocity; S12: Clarify the voltage vector distribution characteristics of the fundamental and harmonic subspaces. S121: A two-level dual three-phase inverter with a common DC bus outputs 64 basic voltage vectors, including 60 non-zero voltage vectors and 4 zero vectors; the 60 non-zero voltage vectors are then used to determine their values ​​based on... subplane and The projections on the subplane are grouped to obtain group D3 and group D4; The subplane is the stationary coordinate system of the fundamental subspace. The subplane is a harmonic subspace; S122: Using a reconstructed set of virtual voltage vectors, assuming the virtual voltage vectors... The duration of action is Let the duration of action of the voltage vectors in groups D3 and D4 be respectively and The relationship is expressed as: in, This is the bus voltage.

3. The control method according to claim 2, characterized in that: Step S2 includes the following steps: S21: In the fundamental subspace, design a duty cycle modulation strategy based on the principle of minimizing the error vector. set up (OA) is a given reference voltage vector with a phase angle of . According to the phase angle Determine the sector where the reference voltage vector is located; select the effective voltage vector corresponding to the sector. (OB), the effective voltage vector and The included angle of the axis is Then the duty cycle Represented as: in, , , ; , They are respectively shaft and Shaft voltage; Based on the obtained duty cycle The duration of action of the effective voltage vector in the fundamental subspace is obtained: in, It is an electrical cycle; S22: Improved design of deadbeat flux torque controller; The axial flux linkage is represented as: in, It is a permanent magnet flux linkage; Substituting equation (1) into equation (6) yields the deadbeat flux torque controller: For surface-mount permanent magnet synchronous motors = , and The discrete domain expression is shown below: in, , , Given a reference torque, To calculate the torque, Represents the current moment; and The discrete domain expression is shown below: The error vector transformation duty cycle modulation required by the inverse Park transform is obtained. and .

4. The control method according to claim 3, characterized in that: Step S3 includes the following steps: S31: Design an online learning harmonic canceller in the harmonic subspace. The 5th and 7th harmonic currents in the harmonic subspace are mapped to... In the subspace, when transformed to a reference coordinate system rotating at the fundamental frequency, a 6th harmonic current appears; based on the known 6th harmonic current, assume... and As shown below: in, It is an electrical cycle; , These are the cosine and sine harmonic basis functions, respectively; The error between the harmonic signal and the compensation signal over one electrical cycle is given by the following formula: in, It is a 6th harmonic current. and They are respectively and Weighting coefficients; S32: Update weight coefficients in real time using the least mean square algorithm; The learning and updating process of the weight coefficients uses the least mean square algorithm, assuming the objective function is... Its expression is: The weight coefficients are obtained by minimizing the objective function using gradient descent. The gradient of the weight coefficients is expressed as: in, express The gradient at the current moment; Introducing adaptive learning rate The adaptive calculation process for the weighting coefficients is as follows: The output signal of the online learning harmonic canceller is expressed as: 。 5. The control method according to claim 4, characterized in that: In step S4, the selected D1 vector set is used as the effective vector of the harmonic subspace, and the sector is divided based on the reference voltage phase to achieve coordinated modulation of harmonic current suppression and fundamental subspace control. The specific steps are as follows: Selecting the D1 vector set , , , The four voltage vectors serve as the effective voltage vectors. When the harmonic subspace reference voltage is located in sector I, , The durations of the voltage vectors are as follows: in, and They are respectively shaft and Shaft voltage; When the harmonic subspace reference voltage is located in sector III , The durations of action of the voltage vectors are as follows:

6. The control method according to claim 5, characterized in that: In step S5, the effective voltage vector of the fundamental subspace is added together with the effective voltage vector of the harmonic subspace to generate the final switching sequence, which is then output to the dual three-phase inverter to effectively suppress torque pulsation and harmonic current of the dual three-phase permanent magnet synchronous motor. The specific steps are as follows: According to equations (4), (16), and (17), the effective voltage vector duration within one electrical cycle is: Alternatively, according to equations (4), (18), and (19), the effective voltage vector's duration within one electrical cycle can be obtained as follows: Based on the above, a switching sequence is generated to effectively suppress torque ripple and harmonic current in a dual three-phase permanent magnet synchronous motor.

Images