Permanent magnet synchronous motor model-free predictive current control method based on extended control set
Through the extended control set and new current gradient establishment method, the problem of current gradient update stagnation in model-free prediction control is solved, the current ripple and total harmonic distortion is reduced, and the steady-state and dynamic performance of permanent magnet synchronous motors is improved.
Patent Information
- Application Number
- CN202411113571.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-14
- Publication Date
- 2024-11-26
AI Technical Summary
The traditional model-free prediction control method is used to stagnate the current gradient update in permanent magnet synchronous motors, resulting in excessive current ripple, affecting the motor operation stability.
Through discrete space vector modulation technology, the control set is expanded from 7 voltage vectors to 37, and a new current gradient establishment method is proposed, combining delay compensation and candidate vector preselecting to reduce the number of candidate voltage vectors and select the optimal voltage vector to reduce computational burden.
It effectively suppresses current ripple and total harmonic distortion, improves the steady-state and dynamic performance of the control system, and reduces the performance requirements for the microcontroller.
Smart Images

Figure CN119030399A_ABST
Abstract
Description
Technical Field
[0001] The invention relates to the technical field of permanent magnet synchronous motor model predictive control, in particular to a permanent magnet synchronous motor model-free predictive current control method based on an extended control set. Background Art
[0002] Compared with asynchronous motors, permanent magnet synchronous motors have the advantages of high efficiency, high power density, low noise, low heat generation, and high reliability. Therefore, they have been widely used in the fields of ship electric propulsion, electric vehicles, robots, aerospace, etc. The traditional permanent magnet synchronous motor control methods mainly include vector control and direct torque control. However, the adjustment of PI parameters of vector control is cumbersome, and the steady-state performance of direct torque control is poor. Model predictive control predicts the impact of voltage vector action on the future performance of the system according to the prediction model. By traversing all optional voltage vectors, the optimization problem of the system is solved online in a given time domain, and finally the selected optimal voltage vector is applied to the motor. Compared with vector control, model predictive control has a simpler control structure, does not require PI tuning, and can obtain fast torque and current responses. At the same time, corresponding optimization targets can be specially designed for different systems, such as suppressing harmonic currents, improving the quality of load currents, and reducing switching frequency. Compared with direct torque control, model predictive control can obtain a relatively fixed switching frequency and better steady-state performance. Therefore, it has received extensive attention from academia and industry.
[0003] Although traditional model predictive control has the advantages of fast response speed and easy implementation, it uses idealized motor modeling and ignores the changes in motor parameters, resulting in the model being heavily dependent on the accuracy of motor parameters. At the same time, the uncertainty of motor parameters during actual operation will inevitably lead to a decline in system control performance. In order to solve this problem and improve the robustness of model predictive control, a model-free predictive control method has been proposed. However, since the existing model-free predictive control method only uses 7 basic voltage vectors, there is a problem of current gradient update stagnation, resulting in excessive current ripple in this method, which seriously affects the stability of motor operation. In order to solve the above problems, it is urgent to study a model-free predictive control method that suppresses the stagnation of current gradient update. Summary of the invention
[0004] In view of the deficiencies in the above-mentioned technologies, the present invention proposes a model-free predictive current control method for torque pulsation suppression of a permanent magnet synchronous motor with an extended control set; the present invention expands the control set from 7 voltage vectors to 37 voltage vectors through discrete space vector modulation, thereby reducing torque pulsation; in addition, current gradient estimation and voltage vector preselection are performed based on the motor current and the applied voltage vector, which effectively reduces the computational burden of the model-free predictive current control algorithm and can effectively downgrade the requirements of the model predictive control algorithm on the performance of the microcontroller.
[0005] The method for model-free predictive current control of a permanent magnet synchronous motor based on an extended control set comprises the following steps:
[0006] Step 1: In the kth control cycle, obtain the three-phase permanent magnet synchronous motor current i a (k), i b (k), i c (k), electromotive force ψ r 、Back electromotive force e s , stator inductance L s 、Rotor electrical angle θ e and the rotor electrical angular velocity ω e And through transformation, we can get the α and β axis current i in the kth control period α (k) and i β (k);
[0007] Step 2: Use discrete space vector modulation technology to construct an extended control set, expanding the original 7 basic voltage vectors in the model-free predictive current control method to 37;
[0008] Step 3: Use the α and β axis components V of the expanded minimum basic voltage vector α and V β , get the current gradient corresponding to the basic voltage vector and And obtain the predicted current of k+1 control cycle after the voltage vector acts and
[0009] Step 4: Obtain the reference current gradient through delay compensation and according to and The value of determines the quadrant where the candidate voltage vector is located, and then narrows the area where the candidate voltage vector is located, reducing the number of candidate voltage vectors from 37 to 6;
[0010] Step 5: Based on the candidate voltage vector V xy , the corresponding current gradient is calculated Then, according to the cost function, the optimal voltage vector is selected from the candidate voltage vectors, and an inverter control signal is generated.
[0011] Furthermore, the α and β axis components V of the expanded minimum basic voltage vector are used. α and V β , get the current gradient corresponding to the basic voltage vector and And obtain the predicted current of k+1 control cycle after the voltage vector acts and The method is:
[0012] In the case of a control cycle T≤100us, since T is very short, it can be assumed that i s With e s In adjacent control cycles, the α and β axis components V of the expanded minimum basic voltage vector are equivalent. α and V β , calculate the current gradient corresponding to the basic voltage vector according to formula (1) and According to formula (2), the predicted current of k+1 control cycle after the voltage vector is applied is calculated as and The calculation method is as follows:
[0013]
[0014] In the formula, G α =V α T / L α , G β =V β T / L β , Δα=xm,Δβ=yn,T is the control period, L α , L β is the stator inductance, x, y, m, n are the minimum basic voltage vector V α and V β The number of actions in k control cycle and k-1 control cycle.
[0015] Furthermore, in formula (4), only G α and G β It cannot be obtained directly by calculation, so we only need to estimate G α and G β , the predicted current can be obtained; G α and G β It can be calculated as:
[0016]
[0017] Furthermore, in step 2, the extended control set constructed using discrete space vector modulation technology is:
[0018]
[0019] In the formula, t i =T s / N, N is the number of discrete steps, These are the 7 basic voltage vectors that have not been expanded.
[0020] The basic voltage vector in the extended control set can be calculated as:
[0021]
[0022] Where V α =U dc / 9, V β =3U dc / 9,U dc is the DC bus voltage.
[0023] Furthermore, in step 4, and The area where the candidate voltage vector is located is reduced by a multiple of .
[0024] Further, the step 4 includes the following:
[0025] 4.1) Through delay compensation, reference current gradient and It can be calculated as:
[0026]
[0027] In the formula, and are the reference currents of the α-axis and β-axis respectively, and are the α-axis and β-axis currents after delay compensation according to (4);
[0028] 4.2) According to and The value of determines the quadrant in which the candidate voltage vector is located:
[0029] like and Then the candidate voltage vector is in quadrant I;
[0030] like and Then the candidate voltage vector is in quadrant II;
[0031] like and Then the candidate voltage vector is in quadrant III;
[0032] like and Then the candidate voltage vector is in quadrant IV;
[0033] 4.3) In the determined quadrant, according to and Reduce the number of candidate voltage vectors by a factor of:
[0034] like Then the candidate voltage vector region is in the upper and lower regions;
[0035] like Then the candidate voltage vector area is in the left and right area;
[0036] like Then the candidate voltage vector region is in the middle region.
[0037] Further, the step 5 comprises:
[0038] 5.1) According to the candidate voltage vector V xy , the corresponding current gradient is calculated The calculation method is as follows:
[0039]
[0040] 5.2) According to the cost function, the optimal voltage vector is selected from the candidate voltage vectors, wherein the cost function is as shown in formula (8):
[0041]
[0042] Furthermore, the α-axis and β-axis currents i in the kth control cycle are obtained by Clark transformation and Park transformation. α (k) and i β (k).
[0043] The advantages and positive effects of the present invention are:
[0044] The present invention proposes a model-free predictive current control method for a permanent magnet synchronous motor based on an extended control set. An extended control set is constructed by using discrete space vector modulation technology, and the control set is expanded from 7 voltage vectors to 37 voltage vectors. At the same time, a new current gradient establishment method is proposed to avoid the stagnation problem of current gradient update, reduce current ripple and total harmonic distortion, and design a new candidate vector preselection method. Through the candidate vector preselection, the candidate voltage vectors are reduced, thereby reducing the calculation amount of the control system. BRIEF DESCRIPTION OF THE DRAWINGS
[0045] Figure 1 A schematic diagram of a control block diagram provided by an embodiment of the present invention;
[0046] Figure 2 A control flow chart provided for an embodiment of the present invention;
[0047] Figure 3 A candidate voltage vector region division diagram provided by an embodiment of the present invention;
[0048] Figure 4 A current and gradient estimation waveform diagram of a three-phase permanent magnet synchronous motor during steady-state operation provided by an embodiment of the present invention;
[0049] Figure 5 This is a speed and current waveform diagram of a three-phase permanent magnet synchronous motor when the speed changes suddenly provided by an embodiment of the present invention. DETAILED DESCRIPTION
[0050] In order to make the purpose, technical solution and advantages of the implementation of the present invention clearer, the technical solution in the embodiment of the present invention will be described in more detail below with reference to the accompanying drawings in the embodiment of the present invention.
[0051] The embodiments of the present invention are described in detail below with reference to the accompanying drawings.
[0052] Figure 1 The present invention is a schematic diagram of a control block diagram. First, the three-phase current, electromotive force, back electromotive force, stator inductance, rotor electrical angle, and rotor electrical angular velocity of a permanent magnet synchronous motor are collected, and the current is converted to the α and β planes by coordinate transformation; secondly, an extended control set is constructed by using discrete space vector modulation technology, and the original 7 basic voltage vectors in the model-free predictive current control method are expanded to 37; then, the current gradient is estimated according to the motor current and the applied voltage vector, and the vector is pre-selected to select a candidate voltage vector; finally, the optimal voltage vector is selected from the candidate voltage vectors according to the cost function, and an inverter control signal is generated.
[0053] Figure 2 The control flow chart of the embodiment of the present invention includes the following steps:
[0054] Step 1: In the kth control cycle, sample and store the three-phase permanent magnet synchronous motor current i a (k), i b (k), i c (k), electromotive force ψ r 、Back electromotive force e s , stator inductance L s 、Rotor electrical angle θ e and the rotor electrical angular velocity ω e , the α and β axis currents i of the kth control cycle are obtained through Clark transformation and Park transformationα (k) and i β (k);
[0055] Step 2: An extended control set is constructed using discrete space vector modulation technology, which expands the original 7 basic voltage vectors in the model-free predictive current control method to 37. The specific calculation method is as follows:
[0056] Create an extended control set as follows:
[0057]
[0058] In the formula, t i =T s / N; N is the number of discrete steps, The 7 basic voltage vectors are not expanded;
[0059] The basic voltage vector in the extended control set can be calculated as:
[0060] V xy =xV α +yV β (2)
[0061] Where V α =U dc / 9, V β =3U dc / 9,U dc is the DC bus voltage, V α 、V β are the components of the minimum basic voltage vector after expansion on the α and β axes, respectively, and x and y are V α and V β The number of actions in the current control cycle.
[0062] Step 3: When the control period T≤100us, since T is very short, it can be assumed that i s With e s In adjacent control cycles, it is equivalent. According to the expanded basic voltage vector, the α and β axis components V of the expanded minimum basic voltage vector are used. α and V β , calculate the current gradient corresponding to the basic voltage vector according to formula (3) and Then use formula (3) to get the current gradient and According to formula (4), the predicted current of k+1 control cycle after the voltage vector is applied is calculated as and The calculation method is as follows:
[0063]
[0064] In the formula, G α =V α T / L α , G β =V β T / L β , Δα=xm and Δβ=yn; T is the control period, L α , L β is the stator inductance, x, y, m, n are the minimum basic voltage vector V α and V β The number of actions in k control cycle and k-1 control cycle;
[0065] In addition, due to L α With L β is unknown, so G in formula (4) α and G β It cannot be directly calculated. However, in formula (4), only G α and G β is unknown, so we only need to estimate G α and G β The predicted current, G α and G β It can be estimated as:
[0066]
[0067] Step 4: Figure 3 As shown, by delay compensation, the reference current gradient is obtained and according to and The value of determines the quadrant in which the candidate voltage vector is located, and then and The area where the candidate voltage vector is located is reduced by a multiple of , and the number of candidate voltage vectors is reduced from 37 to 6. The specific calculation method is as follows:
[0068] 4.1) Through delay compensation, reference current gradient and It can be calculated as:
[0069]
[0070] 4.2) Let V x =xV α , V y =yV β , with V x is the X axis, V y Make a rectangular coordinate system for the Y axis, according to and The value of determines the quadrant in which the candidate voltage vector is located, as shown in the following table:
[0071]
[0072] 4.3) In the determined quadrant, according to and The number of candidate voltage vectors is reduced by a multiple of , as shown in the following table.
[0073]
[0074] Step 5: Based on the candidate voltage vector V xy , the corresponding current gradient is calculated Then, according to the cost function, the optimal voltage vector is selected from the candidate voltage vectors and the inverter control signal is generated. The calculation method is as follows:
[0075] 5.1) According to the candidate voltage vector V xy , the corresponding current gradient is calculated The calculation method is as follows:
[0076]
[0077] 5.2) According to the cost function as shown in formula (8), the optimal voltage vector is selected from the candidate voltage vectors:
[0078]
[0079] In order to verify the effectiveness of the present invention, an experimental verification of a model-free predictive current control method for a permanent magnet synchronous motor based on an extended control set proposed in the present invention was carried out on a 2kw three-phase motor drive experimental platform, with a given torque of 5N·m.
[0080] like Figure 4 The figure shows the current and gradient estimation waveform of the three-phase permanent magnet synchronous motor when the speed is set to 500r / min. Figure 4 It can be seen that the estimated current gradient is constant in steady state, the dq axis current ripples are only 2.1A and 2.2A respectively, and the phase current THD is only 7.2%. The experimental results show the reliability and feasibility of the permanent magnet synchronous motor model-free predictive current control method based on the extended control set proposed in the present invention, and also prove that it effectively suppresses current ripple and total harmonic distortion, and has excellent steady-state performance.
[0081] like Figure 5The figure shows the speed and current waveform of the three-phase permanent magnet synchronous motor when the speed changes suddenly from 500r / min to 1000r / min. It can be seen from the figure that the dynamic adjustment time of this method is 60ms. When the speed changes, the q-axis current can quickly reach its maximum allowable value to provide maximum acceleration. The experimental results show that the proposed method has excellent dynamic performance.
[0082] Finally, it should be pointed out that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit them. Although the present invention has been described in detail with reference to the above embodiments, a person skilled in the art should understand that the technical solutions described in the above embodiments can still be modified, or some of the technical features can be replaced by equivalents; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.
Claims
1. A model-free predictive current control method for a permanent magnet synchronous motor based on an extended control set, characterized in that: The steps include: Step 1: In the kth control cycle, obtain the three-phase permanent magnet synchronous motor current i a (k), i b (k), i c (k), electromotive force ψ r 、Back electromotive force e s , stator inductance L s 、Rotor electrical angle θ e and the rotor electrical angular velocity ω e And through transformation, we can get the α and β axis current i in the kth control period α (k) and i β (k); Step 2: Use discrete space vector modulation technology to construct an extended control set, expanding the original 7 basic voltage vectors in the model-free predictive current control method to 37; Step 3: Use the α and β axis components V of the expanded minimum basic voltage vector α and V β , get the current gradient corresponding to the basic voltage vector and And obtain the predicted current of k+1 control cycle after the voltage vector acts and Step 4: Obtain the reference current gradient through delay compensation and according to and The value of determines the quadrant where the candidate voltage vector is located, and then narrows the area where the candidate voltage vector is located, reducing the number of candidate voltage vectors from 37 to 6; Step 5: Based on the candidate voltage vector V xy , the corresponding current gradient is calculated Then, according to the cost function, the optimal voltage vector is selected from the candidate voltage vectors, and an inverter control signal is generated.
2. The method for model-free predictive current control of a permanent magnet synchronous motor based on an extended control set according to claim 1, characterized in that: The α and β axis components V of the expanded minimum basic voltage vector are used. α and V β , get the current gradient corresponding to the basic voltage vector and And obtain the predicted current of k+1 control cycle after the voltage vector acts and The method is: Assumption i s With e s In adjacent control cycles, the α and β axis components V of the expanded minimum basic voltage vector are equivalent. α and V β , calculate the current gradient corresponding to the basic voltage vector according to formula (1) and According to formula (2), the predicted current of k+1 control cycle after the voltage vector is applied is calculated as and The calculation method is as follows: In the formula, G α =V α T / L α , G β =V β T / L β , Δα=xm,Δβ=yn,T is the control period, L α , L β is the stator inductance, x, y, m, n are the minimum basic voltage vector V α and V β The number of actions in k control cycle and k-1 control cycle.
3. The method for model-free predictive current control of a permanent magnet synchronous motor based on an extended control set according to claim 1, characterized in that: In formula (4), only G α and G β It cannot be obtained directly by calculation, so we only need to estimate G α and G β , the predicted current can be obtained; G α and G β It can be calculated as:
4. The method for model-free predictive current control of a permanent magnet synchronous motor based on an extended control set according to claim 1, characterized in that: In step 2, the extended control set constructed using discrete space vector modulation technology is: Where, t i =T s / N, N is the number of discrete steps, These are the 7 basic voltage vectors that have not been expanded. The basic voltage vector in the extended control set can be calculated as: V xy =xV α +yV β (2) Where V α =U dc / 9, V β =3U dc / 9,U dc is the DC bus voltage.
5. The method for model-free predictive current control of a permanent magnet synchronous motor based on an extended control set according to claim 1, characterized in that: In step 4, and The area where the candidate voltage vector is located is reduced by a multiple of .
6. The method for model-free predictive current control of a permanent magnet synchronous motor based on an extended control set according to claim 1, characterized in that: The step 4 includes the following: 4.1) Through delay compensation, reference current gradient and It can be calculated as: In the formula, and are the reference currents of the α-axis and β-axis respectively, and are the α-axis and β-axis currents after delay compensation according to (4); 4.2) According to and The value of determines the quadrant in which the candidate voltage vector is located: like and Then the candidate voltage vector is in quadrant I; like and Then the candidate voltage vector is in quadrant II; like and Then the candidate voltage vector is in quadrant III; like and Then the candidate voltage vector is in quadrant IV; 4.3) In the determined quadrant, according to and Reduce the number of candidate voltage vectors by a factor of: like Then the candidate voltage vector region is in the upper and lower regions; like Then the candidate voltage vector area is in the left and right area; like Then the candidate voltage vector region is in the middle region.
7. The method for model-free predictive current control of a permanent magnet synchronous motor based on an extended control set according to claim 1, characterized in that: The step 5 comprises: 5.1) According to the candidate voltage vector V xy , the corresponding current gradient is calculated The calculation method is as follows: 5.2) According to the cost function, the optimal voltage vector is selected from the candidate voltage vectors, wherein the cost function is as shown in formula (8):
8. The method for model-free predictive current control of a permanent magnet synchronous motor based on an extended control set according to claim 1, characterized in that: Through Clark transformation and Park transformation, we can get the α and β axis currents i in the kth control cycle. α (k) and i β (k).
Citation Information
Cited By
Surface-mounted permanent magnet motor model-free predictive control method based on current gradient
CN121984402A
Surface-mounted permanent magnet motor model-free predictive control method based on current gradient
CN121984402B
