Direct Torque Control Method for Motors Based on Space Vector Modulation
By employing a space vector modulation direct torque control method in a common DC bus type open-winding permanent magnet synchronous motor, the problems of insufficient voltage vector selection accuracy and the influence of zero-sequence current in traditional methods are solved, achieving precise control of torque and flux linkage, and improving low-speed performance and motor operation stability.
Patent Information
- Application Number
- CN202611368701.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-09-04
- Publication Date
- 2026-11-17
AI Technical Summary
Traditional direct torque control methods in open-winding permanent magnet synchronous motors with a common DC bus suffer from problems such as insufficient voltage vector selection accuracy, large torque and flux pulsation, severe influence of zero-sequence current, and inconsistent switching frequency, resulting in low control accuracy and poor low-speed performance.
A direct torque control method for motors based on space vector modulation is adopted. By dividing the two-phase stationary α and β coordinate system into 12 sectors, the principal vector and zero-sequence suppression vector are selected. Zero-sequence current closed-loop control is performed using a frequency adaptive quasi-resonant controller. A virtual voltage vector is constructed and the duty cycle is calculated. A five-segment space vector modulation is used to generate the switching drive signal.
It significantly improves the accuracy of voltage vector selection, reduces torque and flux pulsation, fixes the switching frequency, improves low-speed operation performance, and achieves precise torque regulation and smooth motor operation.
Smart Images

Figure CN123122399A_ABST
Abstract
Description
Technical Field
[0001] This application relates to a direct torque control method for a motor based on space vector modulation, an electronic device, and a computer-readable storage medium, belonging to the field of control technology for open-winding permanent magnet synchronous motors with a common DC bus. Background Technology
[0002] The open-winding permanent magnet synchronous motor (OW-PMSM) with common DC bus has advantages such as high power density, high efficiency, low hardware cost, and high reliability, and has been widely studied in fields such as electric vehicles, aerospace, and ship propulsion.
[0003] Direct torque control (DTC), as a high-performance AC motor control strategy, offers advantages such as fast dynamic response, simple structure, and low dependence on motor parameters by directly controlling stator flux linkage and electromagnetic torque. However, applying traditional DTC to OW-PMSMs with a common DC bus results in the following drawbacks: Firstly, traditional direct torque control divides the two-phase stationary α and β coordinate planes into six sectors, each covering a 60° electrical angle. The sector division is relatively coarse, resulting in insufficient accuracy in voltage vector selection and low control precision.
[0004] Secondly, traditional direct torque control can only select a single discrete voltage vector to act on the entire control cycle in each control cycle, which cannot achieve more precise vector adjustment, resulting in large torque ripple and flux linkage ripple.
[0005] Third, the OW-PMSM with a common DC bus has a zero-sequence circuit. The third harmonic of the stator flux linkage generates a zero-sequence voltage, which in turn induces a zero-sequence circulating current. Traditional methods, to avoid the influence of zero-sequence current, typically use only the middle vector with a zero-sequence voltage value of zero in the vector lookup table. However, this cannot suppress the impact of the third harmonic of the flux linkage on the motor's operating state. Another method is to suppress the zero-sequence voltage by inserting a zero vector, but the zero vector does not bear the responsibility of exciting electromagnetic torque, and its insertion will also cause torque pulsation, affecting the motor's operating performance.
[0006] Fourth, the switching frequency of traditional direct torque control is not fixed, which makes filter design and system optimization difficult, and its low-speed operation performance is poor. Summary of the Invention
[0007] This application discloses a direct torque control method for motors based on space vector modulation, an electronic device, and a computer-readable storage medium.
[0008] The direct torque control method for motors based on space vector modulation in the embodiments of this application includes: Step 1, Sector Division and Voltage Vector Selection: Divide the two-phase stationary α and β coordinate system into 12 sectors, each sector covering 30° electrical angle; Based on the sector where the stator flux is located obtained by the flux linkage observer, the flux linkage command value output by the flux linkage hysteresis controller, the torque command value output by the torque hysteresis controller, and the polarity of the reference zero-sequence voltage, simultaneously select the main vector and the zero-sequence suppression vector from the reconstructed 12-sector vector lookup table; Step 2, Zero-sequence vector duty cycle calculation: A frequency adaptive quasi-resonant controller is used to perform closed-loop control of the zero-sequence current. The resonant frequency of the frequency adaptive quasi-resonant controller is adjusted in real time with the rotor electric angular velocity, and the reference zero-sequence voltage value is output. Based on the zero-sequence voltage amplitude of the selected zero-sequence suppression vector, the action time and duty cycle of the zero-sequence suppression vector in one control cycle are calculated using the volt-second theorem. Step 3: Calculation of the main vector duty cycle: Based on the deviation between the torque setpoint and the actual value of the sampled electromagnetic torque, construct a five-segment virtual voltage vector along the q-axis, so that the torque increment generated by the virtual voltage vector within one control cycle is equal to the torque deviation; use the spatial vector synthesis relationship between the main vector, the zero-sequence suppression vector and the virtual voltage vector to solve for the main vector duty cycle. Step 4, SVM modulation: Input the main vector and zero-sequence suppression vector selected in step 1, as well as the corresponding duty cycles obtained in steps 2 and 3, into the five-segment space vector modulation module to generate the on-time of each switch of the dual inverter and output a stable switching drive signal.
[0009] In some implementations, the 12-sector vector lookup table in step 1 is constructed as follows: each sector contains six small vectors, two medium vectors, and one large vector, distinguished by their voltage amplitudes; the six small vectors have both positive and negative zero-sequence polarities, the two medium vectors have zero-sequence voltage values, and the large vector has a fixed zero-sequence polarity; the vector lookup table uses the medium vectors as the master vectors and the small vectors that have the same effect on torque and flux linkage as the master vectors as zero-sequence suppression vectors; when the reference zero-sequence voltage polarity is positive, the small vector with a positive zero-sequence voltage is selected; when the reference zero-sequence voltage polarity is negative, the small vector with a negative zero-sequence voltage is selected; and when the reference zero-sequence voltage is zero, the medium vector with a zero-sequence voltage is selected.
[0010] In some implementations, when selecting the zero-sequence suppression vector in step 1, the smallest vector that minimizes the number of switching actions of the dual inverter switching transistors within one control cycle is selected from among the candidate small vectors that meet the zero-sequence polarity requirement as the zero-sequence suppression vector.
[0011] In some implementations, the transfer function of the frequency-adaptive quasi-resonant controller described in step 2 is:
[0012] in For transfer functions, This is the proportionality coefficient. The resonance coefficient, The rotor's electric angular velocity is given, and the resonant frequency is set to 3. To track the third harmonic characteristic of zero-sequence current, The variable is a Laplace variable; the frequency adaptive quasi-resonant controller takes the difference between the zero-sequence current setpoint and the sampled zero-sequence current value as input and outputs the reference zero-sequence voltage value.
[0013] In some implementations, the virtual voltage vector in step 3 is constructed by differentiating and discretizing the electromagnetic torque formula, and setting the torque increment within one control cycle. ,in For the torque setpoint, To sample the electromagnetic torque value, the components of the virtual voltage vector on the q-axis satisfy the following:
[0014] in For virtual voltage vectors, For stator inductance, For extreme logarithms, This refers to the flux linkage amplitude of the rotor permanent magnet. The amplitude of the third harmonic flux linkage of the rotor permanent magnet. It is a change over time. It is the zero-sequence current. The rotor electrical angle; The virtual voltage vector along the q-axis, expressed in the stationary α, β coordinate system, is as follows:
[0015] in This represents the component of the virtual voltage vector along the q-axis. The imaginary unit, α is the angle between the d-axis and the α-axis.
[0016] In some implementations, the formula for calculating the duty cycle of the principal vector in step 3 is:
[0017] in The duty cycle of the zero-order suppression vector. The duty cycle of the main vector. For virtual voltage vectors, The magnitude of the main vector. The angle between the virtual voltage vector and the principal vector. To control the cycle.
[0018] In some implementations, the five-segment space vector modulation module described in step 4 operates as follows: within one control cycle, the action time of each vector is allocated in a symmetrical order of zero vector—main vector—zero-sequence suppression vector—main vector—zero vector; the action time of the zero vector is determined according to the switching state of the dual inverters corresponding to each vector, so as to generate PWM drive signals for the six bridge arms and a total of twelve switching transistors.
[0019] The electronic device in this application includes a memory and a processor. The memory stores a computer program, and when the computer program is executed by the processor, the direct torque control method for motors based on space vector modulation in the above-described embodiments is implemented.
[0020] The computer-readable storage medium in the embodiments of this application stores a computer program that, when executed by one or more processors, implements the space vector modulation-based direct torque control method for motors described above.
[0021] The beneficial effects of this application are as follows: The direct torque control method for motors based on space vector modulation in the embodiments of this application subdivides the α and β planes into 12 sectors, which significantly improves the voltage vector selection accuracy and reduces the control error caused by the large sector span. In each control cycle, the main vector and the zero-sequence suppression vector are selected simultaneously, and the multi-vector synergy is achieved through duty cycle modulation, which effectively reduces torque ripple and flux linkage ripple. In addition, this application constructs a virtual voltage vector on the q-axis and calculates the duty cycle of the main vector using vector synthesis relationship, which makes the torque regulation more precise and avoids the torque ripple problem caused by zero vector insertion in traditional methods. Moreover, this application adopts five-segment space vector modulation, fixes the switching frequency, and improves low-speed operation performance. Attached Figure Description
[0022] Figure 1 This is a schematic diagram of the circuit topology of the common DC bus type open winding permanent magnet synchronous motor in the embodiments of this application; Figure 2 This is a distribution diagram of the voltage vector of the dual inverter system in the α and β planes according to the embodiments of this application; Figure 3 This is a schematic diagram of the 12-sector division in the embodiments of this application; Figure 4 This is a schematic diagram illustrating the spatial relationship between the virtual voltage vector, the zero-sequence suppression vector, and the principal vector in the embodiments of this application; Figure 5 This is an overall framework diagram of the improved SVM direct torque control system in the embodiments of this application; Figure 6 This is a waveform diagram of the switching transistors of the dual inverter under different zero-sequence polarities in the embodiments of this application; Figure 7 This is a torque ripple waveform diagram of the motor direct torque control method based on space vector modulation in the embodiments of this application under steady-state operation; Figure 8 This is a dynamic response waveform diagram of torque pulsation when the load torque changes abruptly, based on the direct torque control method for motors based on space vector modulation in the embodiments of this application. Figure 9 This is a waveform diagram of the dynamic response of the motor speed when the given speed changes abruptly, based on the direct torque control method for motors based on space vector modulation in the embodiments of this application. Figure 10 This is a waveform diagram of the zero-sequence current suppression effect of the motor direct torque control method based on space vector modulation in the embodiments of this application under steady-state operation; Figure 11 This is a waveform diagram of the three-phase current effect of the motor direct torque control method based on space vector modulation under steady-state operation in the embodiments of this application. Detailed Implementation
[0023] The direct torque control method for motors based on space vector modulation in this application includes the following steps: Step 1: Sector division and voltage vector selection: Divide the two-phase stationary α and β coordinate system into 12 sectors, each sector covering 30° electrical angle; Based on the sector where the stator flux is located obtained by the flux linkage observer, the flux linkage command value output by the flux linkage hysteresis controller, the torque command value output by the torque hysteresis controller, and the polarity of the reference zero-sequence voltage, simultaneously select the main vector and the zero-sequence suppression vector from the reconstructed 12-sector vector lookup table.
[0024] Specifically, the space vector modulation-based direct torque control method for motors in this application is applied to a common DC bus type open-winding permanent magnet synchronous motor, the circuit topology of which is as follows: Figure 1 As shown, the two ends of the motor stator winding are connected to inverter 1 (2Level-VSI 1) and inverter 2 (2Level-VSI 2) respectively, and the two inverters share a single DC bus voltage. The phase voltage of the OW-PMSM is the difference between the input voltages of the two inverters, as shown in Formula 1: ... Formula 1 in These represent the A, B, and C phase voltage values of inverter 1 and inverter 2, respectively. These are the upper arm conduction signals for inverter 1 and inverter 2, respectively. The two inverters use a complementary conduction method; when the upper arm is turned on, it means... The lower bridge arm is shut off.
[0025] In the two-phase synchronously rotating dq coordinate system, the stator voltage equation is shown in Equation 2: ... Formula 2 in These represent the d-axis voltage, q-axis voltage, and zero-sequence voltage, respectively. These represent the d-axis current, q-axis current, and zero-sequence current, respectively. For stator resistance, For runtime, These are the d-axis inductance, q-axis inductance, and zero-sequence inductance, respectively. The rotor's electric angular velocity, This refers to the flux linkage amplitude of the rotor permanent magnet. The rotor electrical angle is given.
[0026] Zero-sequence voltage and zero-sequence current As shown in Formula 3.
[0027] ... Formula 3 in These are the A-phase voltage, B-phase voltage, and C-phase voltage of a common DC bus type open-winding permanent magnet synchronous motor. These are the A-phase current, B-phase current, and C-phase current of a common DC bus type open-winding permanent magnet synchronous motor. The dual-inverter system formed by two sets of inverters in a common DC bus type open-winding permanent magnet synchronous motor has a total of 64 switching states, which are distributed in the α and β planes after voltage vector synthesis as follows: Figure 2 As shown. Voltage vectors are classified into four categories based on their voltage amplitude: small vectors at the vertices of the inner hexagon, whose amplitude is... The median vector of the vertices of the hexagon has an amplitude of The large vector at the vertex of the outer hexagon has an amplitude of The origin zero vector has an amplitude of 0. Each vector has a different zero-sequence polarity, and the zero-sequence voltage value can be taken as... , Or 0.
[0028] Regarding sector division and voltage vector selection, in some implementations, the 12-sector vector lookup table in step 1 is constructed as follows: each sector contains six small vectors, two medium vectors, and one large vector, distinguished by their voltage amplitudes; the six small vectors have both positive and negative zero-sequence polarities, the two medium vectors have zero-sequence voltage values, and the large vector has a fixed zero-sequence polarity; the vector lookup table uses the medium vectors as the main vectors and the small vectors that have the same effect on torque and flux linkage as the main vectors as zero-sequence suppression vectors; when the reference zero-sequence voltage polarity is positive, the small vector with a positive zero-sequence voltage is selected; when the reference zero-sequence voltage polarity is negative, the small vector with a negative zero-sequence voltage is selected; and when the reference zero-sequence voltage is zero, the medium vector with a zero-sequence voltage is selected.
[0029] Specifically, such as Figure 3 As shown, the α and β planes are divided into 12 sectors, with sector 1 covering (0°, 30°). Each sector spans 30°, and so on. Each sector contains six small vectors, two medium vectors, and one large vector. The six small vectors have both positive and negative zero-sequence polarities, the medium vectors have zero zero-sequence voltage, and the large vector has a fixed zero-sequence polarity. Because the large vector has a single zero-sequence polarity, it is not included in the vector lookup table.
[0030] A torque hysteresis controller and a flux hysteresis controller are used. Based on the current sector of the stator flux linkage, the torque command value τ, the flux linkage command value φ, and the polarity of the reference zero-sequence voltage, the optimal vector combination is selected from the reconstructed 12-sector vector lookup table. The vector lookup table is shown in Table 1.
[0031] Table 1 Vector Lookup Table
[0032] In some implementations, in step 1, when selecting the zero-sequence suppression vector, among the candidate small vectors that meet the zero-sequence polarity requirement, the vector that minimizes the number of switching actions of the dual inverter switching transistors in one control cycle is preferentially selected to reduce switching losses.
[0033] Specifically, for example, when the torque command value τ=1 and the flux linkage command value φ=1 (both torque and flux linkage need to be increased simultaneously), the vector is selected in sector 1. V 35 As the principal vector; when the reference zero-sequence polarity is positive, a smaller vector is selected. V 85 (zero-sequence voltage is) As the zero-sequence suppression vector, the switching transistor changes only once; when the reference zero-sequence polarity is negative, a smaller vector is selected. V 75 (zero-sequence voltage is) As the zero-sequence suppression vector, the switching transistor changes only once; when the reference zero-sequence polarity is zero, the middle vector is directly selected.V 35 (Zero-sequence voltage is 0). The remaining sectors and command combinations follow the same pattern.
[0034] The direct torque control method for motors based on space vector modulation in the embodiments of this application further includes: Step 2: Zero-sequence vector duty cycle calculation: A frequency adaptive quasi-resonant controller is used to perform closed-loop control of the zero-sequence current. The resonant frequency of the frequency adaptive quasi-resonant controller is adjusted in real time with the rotor electric angular velocity, and the reference zero-sequence voltage value is output. Based on the zero-sequence voltage amplitude of the selected zero-sequence suppression vector, the action time and duty cycle of the zero-sequence suppression vector in one control cycle are calculated using the volt-second theorem.
[0035] Specifically, the above process is implemented as shown in Equation 4, which represents the transfer function of the frequency adaptive quasi-resonant controller (FA-PR controller).
[0036] ... Formula 4 in For transfer functions, This is the proportionality coefficient. The resonance coefficient, The rotor's electric angular velocity is given, and the resonant frequency is set to 3. To track the third harmonic characteristic of zero-sequence current, This is a Laplace variable. Thus, the frequency-adaptive quasi-resonant controller uses the difference between the zero-sequence current setpoint and the sampled zero-sequence current value as input, and outputs a reference zero-sequence voltage value.
[0037] Furthermore, after obtaining the reference zero-sequence voltage value, the action time of the selected zero-sequence suppression vector is calculated using the volt-second theorem based on its zero-sequence voltage amplitude, and then the duty cycle of the zero-sequence vector is calculated. Specifically, in some embodiments, when the zero-sequence voltage amplitude of the selected zero-sequence suppression vector is... At that time, its duration of action is When the zero-sequence voltage amplitude of the selected zero-sequence suppression vector is At that time, its duration of action is ;in For reference zero-sequence voltage value, To control the cycle, The DC bus voltage is used, while the zero-sequence vector duty cycle is used. .
[0038] The direct torque control method for motors based on space vector modulation in the embodiments of this application further includes: Step 3: Calculation of main vector duty cycle: Based on the deviation between the torque setpoint and the actual value of the sampled electromagnetic torque, construct a five-segment virtual voltage vector along the q-axis, so that the torque increment generated by the virtual voltage vector within one control cycle is equal to the torque deviation; use the spatial vector synthesis relationship between the main vector, the zero-sequence suppression vector and the virtual voltage vector to solve for the main vector duty cycle.
[0039] Specifically, the electromagnetic torque of the motor is only related to the component of the stator flux linkage on the q-axis. Therefore, when the main vector duty cycle is set, the torque formula is differentiated and discretized to determine the torque increment within one control cycle. ,in For the torque setpoint, To sample the electromagnetic torque value, the condition satisfied by the virtual voltage vector component on the q-axis is shown in Equation 5.
[0040] ... Formula 5 in For virtual voltage vectors, For stator inductance, For extreme logarithms, This refers to the flux linkage amplitude of the rotor permanent magnet. The amplitude of the third harmonic flux linkage of the rotor permanent magnet. Due to changes over time, It is the zero-sequence current. The rotor electrical angle is given.
[0041] Furthermore, the expression for the virtual voltage vector along the q-axis in the stationary α, β coordinate system is shown in Equation 6.
[0042] ... Formula 6 in This represents the component of the virtual voltage vector along the q-axis. The imaginary unit, α is the angle between the d-axis and the α-axis.
[0043] Based on the above implementation method, a vector synthesis method is further adopted, utilizing the zero-order suppression vector. and principal vector Together, we construct a virtual voltage vector along the q-axis. .set up The duty cycle of the zero-order suppression vector. The duty cycle of the main vector. The angle between the virtual voltage vector and the principal vector. Let the angle between the virtual voltage vector and the zero-sequence suppression vector be the angle between them. Then, after subtracting the projection component of the zero-sequence suppression vector along its direction from the virtual voltage vector, the remaining part is the required principal vector. The q-axis components involved in the synthesis are shown in Formula 7.
[0044] ... Formula 7 Given that the duty cycle of the zero-order suppression vector is known, the remaining part mentioned above (i.e. As is also known, the duty cycle of the principal vector can be determined by using the inverse process of vector composition, as shown in Formula 8.
[0045] ... Formula 8 in The duty cycle of the zero-order suppression vector. The duty cycle of the main vector. For virtual voltage vectors, The magnitude of the main vector. The angle between the virtual voltage vector and the principal vector. To control the cycle.
[0046] The direct torque control method for motors based on space vector modulation in the embodiments of this application further includes: Step 4, SVM modulation: Input the main vector and zero-sequence suppression vector selected in step 1, as well as the corresponding duty cycles obtained in steps 2 and 3, into the five-segment space vector modulation module to generate the on-time of each switch of the dual inverter and output a stable switching drive signal.
[0047] Specifically, the actual operation of the five-segment space vector modulation module is as follows: within one control cycle, the duration of each vector is allocated in a symmetrical order of zero vector—main vector—zero-sequence suppression vector—main vector—zero vector; the duration of the zero vector is determined according to the switching state of the dual inverters corresponding to each vector, so as to generate PWM drive signals for the twelve switching transistors of the six bridge arms. Specifically, the duration of the zero vector is shown in Formula 9.
[0048] ... Formula 9 in This indicates the duration of action of the zero vector.
[0049] For example, taking the vectors shown in Table 2, take V 35 Main vector V 85 Zero-order suppression vector V 85 Taking zero-sequence polarity as an example, the upper bridge arm of inverter 1 S 11 , S 13 , S 15The conduction time is respectively based on V 35 and V 85 The corresponding switching state is determined by the duty cycle; when the zero-sequence polarity is negative, the zero-sequence suppression vector takes... V 75 At the same time, the conduction time of each switch is adjusted accordingly. The five-segment symmetrical modulation ensures that the switching action is evenly distributed, effectively reducing current ripple.
[0050] The overall process of the control strategy described in this embodiment is as follows: Figure 5 As shown: A given torque value is obtained from the speed loop and flux linkage observer. Based on the reference flux linkage value, the torque command value τ and flux linkage command value φ are output through the torque hysteresis controller and flux linkage hysteresis controller. According to the sector where the current stator flux linkage vector is located and the polarity of the reference zero-sequence voltage, the optimal vector combination is output from the 12-sector vector lookup table. Using the difference between the reference torque value and the sampled electromagnetic torque value, as well as the sampled zero-sequence current value, the zero-sequence vector duty cycle is obtained through the FA-PR controller and the virtual vector calculation module. and principal vector duty cycle The vector combination and duty cycle are input into the five-segment space vector modulation module (SVM module), which ultimately generates dual inverter switching drive signals to excite the motor to run. (See diagram) This represents the estimated flux linkage value. This indicates the reference flux linkage value mentioned above. , This represents the dq-axis current after coordinate transformation. This represents the zero-axis current obtained from sampling. Same as in the above embodiments T0 corresponds to the duty cycle of the zero-order suppression vector in the above embodiments. T1 corresponds to the main vector duty cycle in the above embodiments. .
[0051] For example, by outputting the optimal vector combination from the vector lookup table, and knowing the duty cycles of the zero-sequence vector and the principal vector, the conduction time of each switching transistor per unit cycle can be obtained through five-segment modulation. This allows us to obtain, for instance, the optimal vector combination. Figure 6 The main vector shown V 35 `, zero-order vector V 85 `, V 75 For example, the switching waveforms of the transistor under different zero-sequence polarities.
[0052] like Figure 7 As shown, the electromagnetic torque during steady-state operation of the method in this application The torque ripple fluctuates around a given value of 5 N·m, with a peak-to-peak value of approximately 2.8 N·m and a fluctuation range of 4.0 N·m to 6.8 N·m. Compared with traditional direct torque control methods, this application improves the voltage vector selection accuracy through 12-sector subdivision and employs a duty cycle modulation method based on the q-axis virtual voltage vector. This ensures that the effective vector only acts on a portion of the control cycle, avoiding torque overshoot caused by a single vector acting on the entire cycle. This effectively reduces torque ripple, resulting in a smoother electromagnetic torque waveform, thus verifying the superiority of this application in reducing torque ripple.
[0053] like Figure 8 The figure shows the dynamic response waveform of the electromagnetic torque under sudden load torque changes. Before the load torque increases by a step, the electromagnetic torque fluctuates stably around the given value with small torque ripple. When the load torque changes by a step, the electromagnetic torque can respond quickly and track the new given value rapidly. There is only a slight overshoot during the dynamic process, and the adjustment time is short. Subsequently, it enters a new steady-state operation stage, and the torque fluctuates stably around the new given value. Experimental results show that this application has a fast dynamic response capability and a small overshoot when the load torque changes suddenly. This is due to the inherent fast torque response characteristics of direct torque control itself, and the fact that this application effectively suppresses torque ripple while ensuring dynamic response speed through 12-sector subdivision and duty cycle modulation.
[0054] like Figure 9 The figure shows the dynamic response waveform of the motor speed under sudden changes in the speed setpoint. Before the speed setpoint increases by a step, the motor speed operates stably near the setpoint with minimal speed fluctuations. When the speed setpoint changes by a step, the motor speed rises rapidly, with a small overshoot during the dynamic process, and can stabilize at the new speed setpoint in a short time, subsequently entering a new steady-state operation phase. Experimental results show that this application has the characteristics of fast response speed, small overshoot, and high steady-state accuracy under sudden changes in speed setpoint. The coordinated control performance of the speed loop and torque loop is good, verifying the effectiveness and reliability of this application over a wide speed range.
[0055] like Figure 10 As shown, the zero-sequence current during steady-state operation in this application The zero-sequence current fluctuates near zero, with a peak-to-peak value of approximately 4.5A and a fluctuation range of -2.0A to +2.5A. The DC component of the zero-sequence current is effectively suppressed to near zero. In a common DC bus type open-winding permanent magnet synchronous motor, the third harmonic flux linkage of the rotor permanent magnet induces a third-harmonic zero-sequence voltage, which in turn triggers a zero-sequence circulating current. This application uses a frequency FA-PR controller for closed-loop control of the zero-sequence current, and its resonant frequency is adjusted in real time with the rotor's electrical angular velocity to 3... It can accurately track and suppress the third harmonic zero-sequence current across the entire speed range. Simultaneously, by selecting a small vector with corresponding zero-sequence polarity as the zero-sequence suppression vector through a vector lookup table and calculating its duty cycle using the volt-second theorem, active compensation of the zero-sequence voltage is achieved. Experimental results show that this application can effectively suppress the zero-sequence circulating current under steady-state operation, maintaining the zero-sequence current amplitude at a low level.
[0056] like Figure 11 As shown, the three-phase current during steady-state operation of the method in this application is... The three-phase current waveform exhibits a good sinusoidal waveform with an amplitude of approximately 15A, a phase difference of 120° electrical angle, and good symmetry, without significant distortion or third harmonic components. This is because the FA-PR controller effectively suppresses zero-sequence circulating current, reducing the coupling effect of zero-sequence current in the three-phase windings. Simultaneously, the five-segment space vector modulation evenly distributes switching actions across the first and second halves of the control cycle, reducing current ripple and fixing the switching frequency. The three-phase sinusoidal current waveform verifies that the method described in this application can drive the motor to operate smoothly with low current harmonic content and good motor performance.
[0057] comprehensive Figures 7 to 11 The experimental results show that the improved SVM direct torque control method described in this application can simultaneously achieve torque ripple suppression, zero-sequence circulating current suppression, and three-phase sinusoidal current output under steady-state operation. It has fast dynamic response capability and small overshoot when there are sudden changes in load torque and speed setting. This verifies the effectiveness and feasibility of the technical solution in this application in terms of steady-state performance and dynamic performance.
[0058] The electronic device in this application includes a memory and a processor. The memory stores a computer program. When the computer program is executed by the processor, the direct torque control method for motors based on space vector modulation in the above-described embodiments is implemented.
[0059] The computer-readable storage medium in the embodiments of this application stores a computer program that, when executed by one or more processors, implements the motor direct torque control method based on space vector modulation in the above embodiments.
[0060] The beneficial effects of this application are as follows: The direct torque control method for motors based on space vector modulation in the embodiments of this application subdivides the α and β planes into 12 sectors, which significantly improves the voltage vector selection accuracy and reduces the control error caused by the large sector span. In each control cycle, the main vector and the zero-sequence suppression vector are selected simultaneously, and the multi-vector synergy is achieved through duty cycle modulation, which effectively reduces torque ripple and flux linkage ripple. In addition, this application constructs a virtual voltage vector on the q-axis and calculates the duty cycle of the main vector using vector synthesis relationship, which makes the torque regulation more precise and avoids the torque ripple problem caused by zero vector insertion in traditional methods. Moreover, this application adopts five-segment space vector modulation, fixes the switching frequency, and improves low-speed operation performance.
[0061] The above description is merely a preferred embodiment of this application and is not intended to limit this application in any way. Although this application has disclosed the preferred embodiment as above, it is not intended to limit this application. Any person skilled in the art can make some modifications or alterations to the above-disclosed technical content to create equivalent embodiments without departing from the scope of the technical solution of this application. Any simple modifications, equivalent substitutions, and improvements made to the above embodiments without departing from the technical solution of this application, based on the technical essence of this application and within the spirit and principles of this application, shall still fall within the protection scope of the technical solution of this application.
Claims
1. A method of direct torque control of an electric machine based on space vector modulation, characterized in that, The method is used to control a common DC bus type open-winding permanent magnet synchronous motor, and the method includes: Step 1, Sector Division and Voltage Vector Selection: Divide the two-phase stationary α and β coordinate system into 12 sectors, each sector covering 30° electrical angle; Based on the sector where the stator flux is located obtained by the flux linkage observer, the flux linkage command value output by the flux linkage hysteresis controller, the torque command value output by the torque hysteresis controller, and the polarity of the reference zero-sequence voltage, simultaneously select the main vector and the zero-sequence suppression vector from the reconstructed 12-sector vector lookup table; Step 2, Zero-sequence vector duty cycle calculation: A frequency adaptive quasi-resonant controller is used to perform closed-loop control of the zero-sequence current. The resonant frequency of the frequency adaptive quasi-resonant controller is adjusted in real time with the rotor electric angular velocity, and the reference zero-sequence voltage value is output. Based on the zero-sequence voltage amplitude of the selected zero-sequence suppression vector, the action time and duty cycle of the zero-sequence suppression vector in one control cycle are calculated using the volt-second theorem. Step 3: Calculation of the main vector duty cycle: Based on the deviation between the torque setpoint and the actual value of the sampled electromagnetic torque, construct a five-segment virtual voltage vector along the q-axis, so that the torque increment generated by the virtual voltage vector within one control cycle is equal to the torque deviation; use the spatial vector synthesis relationship between the main vector, the zero-sequence suppression vector and the virtual voltage vector to solve for the main vector duty cycle. Step 4, SVM modulation: Input the main vector and zero-sequence suppression vector selected in step 1, as well as the corresponding duty cycles obtained in steps 2 and 3, into the five-segment space vector modulation module to generate the on-time of each switch of the dual inverter and output a stable switching drive signal.
2. The method according to claim 1, characterized in that, The 12-sector vector lookup table described in step 1 is constructed as follows: each sector contains six small vectors, two medium vectors, and one large vector, distinguished by their voltage amplitudes; the six small vectors have both positive and negative zero-sequence polarities, the two medium vectors have zero-sequence voltage values, and the large vector has a fixed zero-sequence polarity; the vector lookup table uses the medium vectors as the main vectors and the small vectors that have the same effect on torque and flux linkage as the main vectors as zero-sequence suppression vectors. When the reference zero-sequence voltage polarity is positive, a small vector with positive zero-sequence voltage is selected; when the reference zero-sequence voltage polarity is negative, a small vector with negative zero-sequence voltage is selected; when the reference zero-sequence voltage is zero, a medium vector with zero zero-sequence voltage is selected.
3. The method according to claim 2, characterized in that, When selecting the zero-sequence suppression vector in step 1, among the candidate small vectors that meet the zero-sequence polarity requirement, the small vector that minimizes the number of switching actions of the dual inverter switching transistors in one control cycle is selected as the zero-sequence suppression vector.
4. The method according to claim 1, characterized in that, The transfer function of the frequency adaptive quasi-resonant controller described in step 2 is: in For transfer functions, This is the proportionality coefficient. The resonance coefficient, The rotor's electric angular velocity is given, and the resonant frequency is set to 3. To track the third harmonic characteristic of zero-sequence current, The variable is a Laplace variable; the frequency adaptive quasi-resonant controller takes the difference between the zero-sequence current setpoint and the sampled zero-sequence current value as input and outputs the reference zero-sequence voltage value.
5. The method according to claim 1, characterized in that, The duty cycle of the zero-sequence suppression vector in step 2 is calculated as follows: when the zero-sequence voltage amplitude of the selected zero-sequence suppression vector is... At that time, its duration of action is ; When the zero-sequence voltage amplitude of the selected zero-sequence suppression vector is At that time, its duration of action is ;in For reference zero-sequence voltage value, To control the cycle, DC bus voltage, zero-sequence vector duty cycle .
6. The method according to claim 1, characterized in that, The virtual voltage vector described in step 3 is constructed by differentiating and discretizing the electromagnetic torque formula, and then setting the torque increment within one control cycle. ,in For the torque setpoint, To sample the electromagnetic torque value, the components of the virtual voltage vector on the q-axis satisfy the following: in For virtual voltage vectors, For stator inductance, For extreme logarithms, This refers to the flux linkage amplitude of the rotor permanent magnet. The amplitude of the third harmonic flux linkage of the rotor permanent magnet. It is a change over time. It is the zero-sequence current. The rotor electrical angle; The virtual voltage vector along the q-axis, expressed in the stationary α, β coordinate system, is as follows: in This represents the component of the virtual voltage vector along the q-axis. The imaginary unit, α is the angle between the d-axis and the α-axis.
7. The method according to claim 1, characterized in that, The formula for calculating the duty cycle of the principal vector in step 3 is as follows: in The duty cycle of the zero-order suppression vector. The duty cycle of the main vector. For virtual voltage vectors, The magnitude of the main vector. The angle between the virtual voltage vector and the principal vector. To control the cycle.
8. The method according to claim 1, characterized in that, The five-segment space vector modulation module described in step 4 operates as follows: within one control cycle, the action time of each vector is allocated in a symmetrical order of zero vector—main vector—zero-sequence suppression vector—main vector—zero vector; the action time of the zero vector is determined according to the switching state of the dual inverters corresponding to each vector, so as to generate PWM drive signals for the twelve switching transistors of the six bridge arms.
9. An electronic device, characterized in that, The electronic device includes a memory and a processor. The memory stores a computer program that, when executed by the processor, implements the motor direct torque control method based on space vector modulation as described in any one of claims 1-8.
10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by one or more processors, implements the direct torque control method for motors based on space vector modulation as described in any one of claims 1-8.
