A method, system and medium for suppressing torque ripple in a permanent magnet motor predictive control

Abstract

The application provides a permanent magnet motor predictive control torque ripple suppression method, system and medium, and relates to the technical field of permanent magnet motor control, and comprises the following steps: obtaining two-phase current, rotating speed and rotor position of the permanent magnet motor measured in real time; converting the two-phase current into d, q axis current according to the rotor position, and converting seven groups of alternative voltage vectors of the inverter into seven groups of d, q axis voltage; constructing a deadbeat current prediction model to predict future seven groups of current state values; constructing a sliding mode observer to estimate the load torque; determining the estimated torque ripple value according to the estimated load torque; constructing a cost function; determining the alternative voltage vector corresponding to the minimum value of the cost function in the future seven groups of current state values and the estimated torque ripple value as the optimal voltage vector, and generating a control signal for controlling the permanent magnet motor according to the switching state corresponding to the optimal voltage vector. The method can effectively suppress the torque ripple of the permanent magnet motor.

Description

TECHNICAL FIELD

[0001] The application relates to the technical field of permanent magnet motor control, and particularly relates to a permanent magnet motor predictive control torque ripple suppression method, a system and a medium. BACKGROUND

[0002] As an advanced control strategy, model predictive control is widely used in the control of permanent magnet motors for robots due to its ability to handle multivariable control systems and optimize dynamic performance. Model predictive control maximizes the dynamic response and stability of the motor system by optimizing control inputs in real time. However, due to the existence of torque ripple, model predictive control cannot completely eliminate this fluctuation when applied to permanent magnet motors, especially in fields such as robots that require extremely smooth control. Torque ripple can cause control errors and affect the precision and stability of robots.

[0003] To solve the problem of torque ripple in permanent magnet motors, traditional methods focus on optimizing motor structure design and improving control algorithms. First, structural improvements such as optimizing winding design and adjusting pole shape can effectively reduce torque ripple. For example, Gao Jian et al. in the patent "Permanent Magnet Synchronous Motor Cogging Torque Ripple Suppression Method" changed the structure design of the stator and rotor to reduce the cogging effect, thereby reducing torque ripple. Second, control algorithm optimization is also widely used. For example, Qiu Xin et al. in the patent "Permanent Magnet Synchronous Motor Torque Flux Control Method" introduced torque and flux observers to dynamically compensate torque fluctuations and improve motor output stability. Although these methods alleviate the problem of torque ripple to some extent, due to the limitations of structure and algorithm, the control accuracy in dynamic environments is low, and it is still difficult to meet the needs of precision application scenarios such as robots that require smooth control. SUMMARY

[0004] To solve the above problems, the application provides a permanent magnet motor predictive control torque ripple suppression method. This method is based on a sliding mode observer, estimates the load torque in real time, extracts the torque ripple signal, and integrates it into the cost function of model predictive control, thereby effectively suppressing torque ripple. Compared with traditional methods, this application improves dynamic control accuracy and is particularly suitable for application scenarios such as robots that require smooth control, providing a reliable torque ripple control solution.

[0005] To achieve the above purpose, the application provides the following technical solutions.

[0006] The application provides a permanent magnet motor predictive control torque ripple suppression method, including the following steps:

[0007] acquire real-time measured two-phase current, speed and rotor position of the permanent magnet motor; convert the two-phase current into d, q-axis current according to the rotor position, and convert seven groups of alternative voltage vectors of the inverter into seven groups of d, q-axis voltage;

[0008] construct a dead-beat current prediction model, and predict future seven groups of current state values through the dead-beat current prediction model according to the current motor speed, d, q-axis current and seven groups of d, q-axis voltage;

[0009] construct a sliding mode observer to estimate the speed of the permanent magnet motor, determine the difference between the estimated speed and the real-time measured speed, and determine the estimated load torque according to the difference; determine the estimated torque ripple value according to the estimated load torque;

[0010] construct a cost function with the difference between the actual current and the expected current and the difference between the estimated torque ripple value and the expected torque ripple value; determine the alternative voltage vector corresponding to the minimum value of the cost function in the future seven groups of current state values and the estimated torque ripple value as the optimal voltage vector, and generate a control signal for controlling the permanent magnet motor according to the switching state corresponding to the optimal voltage vector.

[0011] Preferably, the conversion of the two-phase current into d, q-axis current according to the rotor position is as follows:

[0012]

[0013] wherein, i d (k) and i q (k) are d, q-axis current of the permanent magnet motor; i a (k) and i b (k) are a, b two-phase current of the permanent magnet motor; θ(k) is the rotor position.

[0014] Preferably, the conversion of seven groups of alternative voltage vectors of the inverter into seven groups of d, q-axis voltage is as follows:

[0015]

[0016] wherein, u d (k) and u q (k) are d, q-axis voltage; U dc is the system DC bus voltage; [s a , s b , s c ] T is the switching state corresponding to the seven groups of alternative voltage vectors of the inverter, which are [0, 0, 0] T , [1, 0, 0] T , [1, 1, 0] T , [0, 1, 0] T, [0,1,1] T , [0,0,1] T , [1,0,1] T .

[0017] Preferably, the deadbeat current prediction model is as shown in the following formula:

[0018]

[0019] Among them, i d (k+1) and i q (k+1) is the current state value of the next seven groups; i d (k) and i q (k) is the d,q axis current of the permanent magnet motor; ω m (k) is the motor speed; p is the number of motor pole pairs; T s is the control period; ψ f is the permanent magnet flux, R s is the motor resistance, L d and L q are the d-axis and q-axis inductances respectively.

[0020] Preferably, determining the estimated torque ripple value according to the estimated load torque comprises the following steps:

[0021] Construct a sliding mode observer:

[0022]

[0023] Where J is the motor moment of inertia; B is the friction coefficient; k1 is the observer coefficient, which is 4 times the rated torque of the motor. is the estimated speed ω of the sliding mode observer m * (k) and the actual motor speed ω m (k), For the following function:

[0024]

[0025] The motor load torque value T estimated by the sliding mode observer l * (k) is:

[0026]

[0027] The future current state value i d (k+1) and i q (k+1) and the motor load torque value T l * (k) Substitute the following formula to calculate the torque ripple value |T rip(k)|:

[0028] |T rip (k)|=1.5p((L d -L q )i d (k)i q (k)+Ψ f i q (k))-Bω m (k)-T l * (k)。

[0029] Preferably, the cost function is as follows:

[0030]

[0031] Wherein, i d_ref , i q_ref is the current reference value, T rip_ref is the desired torque ripple value.

[0032] The application also provides a permanent magnet motor predictive control torque ripple suppression system, the system comprising:

[0033] a processor;

[0034] a memory having a computer program stored thereon and executable on the processor;

[0035] Wherein, the computer program is executed by the processor to realize the steps of the permanent magnet motor predictive control torque ripple suppression method.

[0036] The application also provides a computer readable storage medium, the computer readable storage medium having a data processing program stored thereon, the data processing program being executed by a processor to realize the steps of the permanent magnet motor predictive control torque ripple suppression method.

[0037] The application has the following advantages:

[0038] The application provides a permanent magnet motor predictive control torque ripple suppression method, system and medium, the application first constructs a sliding mode load torque observer to accurately estimate the load torque. Then, through the difference between the observed load torque and the electromagnetic torque, the torque ripple signal is extracted. Finally, the torque ripple signal is integrated into the cost function of the predictive control, so as to realize effective suppression of the torque ripple while optimizing the control process, especially suitable for precision application scenarios such as robots which have high requirements for smooth control, significantly improving the stability and precision control level of the system. BRIEF DESCRIPTION OF DRAWINGS

[0039] Figure 1is a flow chart of a permanent magnet motor predictive control torque ripple suppression method of an embodiment of the present application.

[0040] Figure 2 is a principle block diagram of a sliding mode observer of an embodiment of the present application. DETAILED DESCRIPTION

[0041] In order to make the purpose, technical solutions and advantages of the present application clearer, the present application is further described in detail below in combination with the drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and do not limit the present application.

[0042] Embodiment 1

[0043] Permanent magnet motors for robots are widely used in model predictive control, but there is a torque ripple problem. The traditional solution mainly reduces the torque ripple by optimizing the motor structure design or improving the control algorithm. However, these methods are limited by the limitations of structure and algorithm, and it is difficult to completely eliminate the torque ripple in high-precision application scenarios. The present application proposes a method for suppressing predictive control torque ripple through a sliding mode observer. First, by constructing a sliding mode observer, the load torque of the motor is accurately estimated in real time, and the actual torque ripple signal is extracted. Second, the signal is integrated into the cost function of model predictive control, and the optimal control voltage vector is dynamically selected by optimizing the control strategy. Finally, using this control strategy, the torque ripple is effectively suppressed, and the dynamic response and control accuracy of the motor system are improved.

[0044] The flow of the method for suppressing predictive control torque ripple through a sliding mode observer proposed by the present application is shown in Figure 1 The electrical parameters of the permanent magnet motor in the present embodiment are as follows: the inductance is L d = L q = 0.3 mH, the resistance is R s = 0.5 Ω, the rotor flux is ψ f = 0.083 Wb, the number of permanent magnets is p = 3, the rated speed is 300 rad / s, the control period is T s = 0.0002 s, the motor moment of inertia J = 0.0001, and the friction coefficient B = 0.0022. The method specifically includes the following steps:

[0045] S1: State measurement. Measure the a, b phase currents of the permanent magnet motor, denoted as i a (k) and i b (k), the motor speed ω m (k) and the rotor position θ(k).

[0046] S2: Coordinate transformation. According to the rotor position θ(k), the phase currents i a (k) and i b(k) to d, q-axis currents i d (k) and i q (k):

[0047]

[0048] S3: Control voltage calculation. According to the rotor position θ(k), the seven sets of alternative voltage vectors of the inverter are converted to seven d, q-axis voltages u d (k) and u q (k):

[0049]

[0050] where U dc is the system DC bus voltage; [s a , s b , s c ] T is the seven sets of alternative voltage vectors of the inverter corresponding to the switching states, which are [0, 0, 0] T , [1, 0, 0] T , [1, 1, 0] T , [0, 1, 0] T , [0, 1, 1] T , [0, 0, 1] T , [1, 0, 1] T .

[0051] S4: Predict the future state of the current. The motor pole pair number p = 3, control period T s = 0.0002, permanent magnet flux ψ f = 0.083, d, q-axis inductance L d = L q = 0.0003, d, q-axis current i d (k) and i q (k), seven sets of d, q-axis control voltages u d (k) and u q (k), motor speed ω m (k) are substituted into the following permanent magnet motor current prediction model:

[0052]

[0053] The future seven sets of current state values i d (k+1) and i q (k+1) can be predicted as:

[0054]

[0055] S5: Estimate the load torque using the sliding mode observer. The schematic diagram of the sliding mode observer is shown in FIG. 3. The motor moment of inertia J = 0.0001, the friction coefficient B = 0.0022, the permanent magnet flux linkage ψ = 0.083, the motor pole pair number p = 3, the observer coefficient k1 = 4, the d, q axis currents i f (k) and i d (k) are substituted into the following sliding mode observer:

[0056]

[0057] The following can be obtained:

[0058]

[0059] wherein, is the estimated speed ω m * (k) of the sliding mode observer, m (k) is the difference between the estimated speed ω l * (k) of the sliding mode observer and the actual speed ω is the following function:

[0060]

[0061] The motor load torque value T d q (k) estimated by the sliding mode observer is:

[0062]

[0063] S6: Extract the torque ripple value. The current future state i l * (k+1) and i rip (k+1) and the estimated load torque T d_ref (k) are substituted into the following formula to calculate the torque ripple value |T q_ref (k)|

[0064]

[0065] S7: Construct the cost function. Considering the current and torque ripple, the following cost function J1 is constructed:

[0066]

[0067] wherein, i rip_ref , i rip_ref is the current reference value, and T rip is the expected torque ripple value.

[0068] S8: Select the optimal voltage vector and generate the control signal ​

[0069] The expected torque ripple value T rip_ref The estimated torque ripple value |T rip (k) and the current seven-group future state i d (k+1) corresponding to the seven groups of alternative voltage vectors, i q (k+1) and the current reference value i d_ref , i q_ref The seven groups of cost function values are compared in size, respectively, and are sorted from small to large. The alternative voltage vector that makes the cost function value minimum is taken as the optimal voltage vector, and a control signal is generated according to the switching state corresponding to the optimal voltage vector, which is used to control the motor.

[0070] The above is the permanent magnet motor predictive control torque ripple suppression method provided by one embodiment of the present embodiment. Based on the same idea, the present embodiment also provides a corresponding permanent magnet motor predictive control torque ripple suppression system. The specific limitations of the permanent magnet motor predictive control torque ripple suppression system can be referred to the limitations of the permanent magnet motor predictive control torque ripple suppression method in the foregoing, and will not be repeated here. Each module in the above permanent magnet motor predictive control torque ripple suppression system can be realized by software, hardware, and combinations thereof, in whole or in part. The above modules can be embedded in or independent of the processor in the computer device in hardware form, or can be stored in the memory in the computer device in software form, so as to be called and executed by the processor to perform the operations corresponding to each of the above modules.

[0071] The present embodiment also provides a computer readable storage medium, which stores a computer program. The computer program can be used to execute the permanent magnet motor predictive control torque ripple suppression method provided by the above Figure 1 The present embodiment also provides a computer readable storage medium, which stores a computer program. The computer program can be used to execute the permanent magnet motor predictive control torque ripple suppression method provided by the above

[0072] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a non-volatile computer readable storage medium, and when executed, can include the processes of the above-mentioned embodiments of the methods. Any reference to memory, storage, database or other medium used in the embodiments provided in the present application can include at least one of non-volatile and volatile memory. The non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory or optical memory, etc. The volatile memory can include random access memory (RAM) or external cache memory. As an illustration but not limitation, the RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM), etc.

[0073] The above only describes the preferred embodiments of the present application and is not intended to limit the present application. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. A method for suppressing torque ripple in predictive control of a permanent magnet motor, characterized in that: The following steps are involved: Obtain the two-phase current, speed and rotor position of the permanent magnet motor in real time; convert the two-phase current into d , q axis current, and convert the seven sets of alternative voltage vectors of the inverter into seven sets d , q Shaft voltage; Construct a deadbeat current prediction model based on the current motor speed, d , q Shaft current and seven groups d , q Shaft voltage, using the deadbeat current prediction model to predict the future seven sets of current state values; Constructing a sliding mode observer to estimate the speed of the permanent magnet motor, determining the difference between the estimated speed and the real-time measured speed, and determining an estimated load torque based on the difference; determining an estimated torque ripple value based on the estimated load torque; A cost function is constructed using the difference between the predicted current and the expected current, as well as the difference between the estimated torque pulsation value and the expected torque pulsation value. The alternative voltage vector corresponding to the minimum value of the cost function among the seven future sets of current state values ​​and estimated torque pulsation values ​​is determined as the optimal voltage vector, and a control signal for controlling the permanent magnet motor is generated according to the switching state corresponding to the optimal voltage vector.

2. The method for suppressing torque ripple in predictive control of a permanent magnet motor according to claim 1, characterized in that: The two-phase current is converted into d , q The shaft current is shown in the following formula: ； in, i d ( k )and i q ( k ) is the permanent magnet motor d , q Shaft current; i a ( k )and i b ( k ) is the permanent magnet motor a , b Two-phase current; θ ( k ) is the rotor position.

3. The method for suppressing torque ripple in predictive control of a permanent magnet motor according to claim 1, characterized in that: The seven groups of alternative voltage vectors of the inverter are converted into seven groups d , q The shaft voltage is given by the following equation: ； in, u d ( k )and u q ( k )for d , q Shaft voltage; U dc is the system DC bus voltage; [ s a , s b , s c ] T are the switching states corresponding to the seven sets of alternative voltage vectors of the inverter, which are [0, 0, 0] T , [1, 0, 0] T , [1, 1, 0] T , [0,1, 0] T , [0, 1, 1] T , [0, 0, 1] T , [1, 0, 1] T .

4. The method for suppressing torque ripple in predictive control of a permanent magnet motor according to claim 1, characterized in that: The deadbeat current prediction model is shown in the following formula: ； in, i d ( k+ 1) and i q ( k+ 1) The seven sets of current state values ​​in the future; i d ( k )and i q ( k ) is the permanent magnet motor d , q Shaft current; ω m ( k ) is the motor speed measured in real time; p is the number of motor pole pairs; T s To control the cycle; ψ f is the permanent magnet flux, R s is the motor resistance, L d and L q They are d , q Shaft inductance.

5. The method for suppressing torque ripple in predictive control of a permanent magnet motor according to claim 1, characterized in that: Determining the estimated torque ripple value according to the estimated load torque comprises the following steps: Construct a sliding mode observer: ； in, J is the motor moment of inertia; B is the friction coefficient; k 1 is the observer coefficient, and its value is 4 times the rated torque of the motor. is the estimated speed of the sliding mode observer ω m * ( k ) and the actual motor speed ω m ( k ), F ( ) is the following function: ； Motor load torque value estimated by sliding mode observer T l * ( k )for: ； The future current state value i d ( k+ 1) and i q ( k+ 1) and motor load torque value T l * ( k ) into the following formula to calculate the torque ripple value | T rip ( k )|: 。 6. The method for suppressing torque ripple in predictive control of a permanent magnet motor according to claim 1, characterized in that: The cost function is shown below: ； in, i d_ref , i q_ref is the current reference value, T rip_ref is the desired torque ripple value.

7. A permanent magnet motor predictive control torque ripple suppression system, characterized in that: The system comprises: processor; a memory having stored thereon a computer program executable on the processor; Wherein, when the computer program is executed by the processor, the steps of the method for suppressing torque pulsation in predictive control of a permanent magnet motor according to any one of claims 1 to 6 are implemented.

8. A computer-readable storage medium, characterized in that The computer-readable storage medium stores a data processing program, which, when executed by a processor, implements the steps of the method for suppressing torque pulsation in predictive control of a permanent magnet motor according to any one of claims 1 to 6.

Images