Three-level dual three-phase PMSM system FCS-MPC method based on multi-target decoupling optimization

Abstract

The invention relates to a three-level dual three-phase PMSM system FCS-MPC method based on multi-target decoupling optimization, and belongs to the technical field of permanent magnet synchronous motor control. Aiming at the problem of incapability of synchronous optimization caused by multi-target coupling in the existing finite set model prediction control, a virtual vector set of low common-mode voltage and zero-harmonic components is constructed to suppress harmonic waves, common-mode voltage and switching frequency, and a balance factor is adjusted by combining a redundant virtual vector and a dead-beat controller to realize active balance of capacitor neutral-point voltage. And the periodic alternate switching frequency is reduced by adopting staggered PWM modulation. According to the invention, accurate tracking of the dq-axis current is realized, the common-mode voltage is reduced to Udc / 12, the phase current THD is lower than 13%, the switching frequency is less than or equal to 4.3 kHz, and the multi-objective optimization performance of the system is significantly improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of permanent magnet synchronous motor control, and relates to a finite control set model predictive control (FCS-MPC) method for a three-level dual three-phase permanent magnet synchronous motor (PMSM) system based on multi-objective decoupling optimization. Background Art

[0002] The dual three-phase permanent magnet electric drive system driven by a T-type three-level has been widely used in high-power transmission fields such as all-electric aircraft, ship propulsion, and new energy vehicles due to its high power density and strong fault tolerance ability. This electric drive system has multiple control degrees of freedom, and the application scenarios are complex. In addition to the two basic control objectives of torque and flux linkage (dq-axis current) and harmonic current (xy-axis current), there are also requirements for low common-mode voltage, capacitor midpoint voltage balance, and low switching frequency. In recent years, the finite set model predictive control has been widely used in the T-type three-level electric drive system due to its simple structure, fast dynamic response, and ability to perform multi-objective optimization using a cost function.

[0003] However, the existing optimization methods embody the influence of the candidate vectors on the sub-objectives in the form of cost sub-functions and integrate them into a unified cost function, or allocate weight coefficients according to the importance degree of the objectives, and use the calculation results of the unified cost function to evaluate each candidate vector. Or use a modified cost function topology (such as the series cost function topology or the trajectory extrapolation type cost function used for sequential model predictive control) to achieve weightless multi-objective optimization. The existing multi-objective optimization schemes optimize the sub-objectives in the form of a cost function, and the candidate vectors simultaneously affect the evaluation results of all sub-objectives, resulting in strong coupling between the sub-objectives and making it difficult to achieve synchronous optimization of all control objectives.

[0004] In view of this, a finite set model predictive control method for a T-type three-level dual three-phase permanent magnet electric drive system based on multi-objective decoupling optimization is proposed. Summary of the Invention

[0005] In view of this, the purpose of the present invention is to provide a finite control set model predictive control (FCS-MPC) method for a three-level dual three-phase permanent magnet synchronous motor (PMSM) system based on multi-objective decoupling optimization. In traditional finite set model predictive control, multi-objective optimization is carried out through a cost function. However, the candidate vectors will affect the evaluation results of all sub-objectives, resulting in coupling with the evaluation results of all sub-objectives and making it difficult to achieve the optimal of all objectives. To better meet the multi-objective optimization requirements in a T-type three-level dual three-phase permanent magnet electric drive system (including five sub-objectives: dq-axis current tracking, harmonic current, common mode voltage, switching frequency suppression, and capacitor midpoint voltage balance), the present invention proposes a finite set model predictive control method for decoupling multi-objectives to achieve synchronous optimization of all sub-objectives.

[0006] This method first realizes the decoupled passive suppression of harmonic current, low common mode voltage, and switching frequency by constructing a virtual vector set with low common mode voltage and zero harmonic voltage components. Secondly, redundant virtual vectors with the same phase and amplitude as the virtual vectors are designed, and the balance factor w that determines the duration of the midpoint current in the redundant virtual vectors is adjusted by a deadbeat controller to achieve the active control of capacitor midpoint voltage balance. In addition, a dynamic vector adjustment scheme is introduced to reduce the number of switchings during period alternation. Through the proposed invention, the dq-axis current of the motor can be accurately tracked, and at the same time, the decoupled synchronous optimization control of harmonic current, common mode voltage, switching frequency suppression, and capacitor midpoint voltage balance can be achieved.

[0007] To achieve the above object, the present invention provides the following technical solutions:

[0008] A finite control set model predictive control (FCS-MPC) method for a three-level dual three-phase permanent magnet synchronous motor (PMSM) system based on multi-objective decoupling optimization, comprising the following steps:

[0009] S1: Construct a virtual vector set with low common mode voltage and zero harmonic voltage components for suppressing harmonic current, common mode voltage, and switching frequency within a control period;

[0010] S2: Traverse the optimal virtual vector through a cost function according to the virtual vector set, and perform delay compensation on the current based on the historical optimal vector;

[0011] S3: Detect the fluctuation of the capacitor midpoint voltage. If it exceeds the preset threshold, start a deadbeat controller to calculate the balance factor w, and replace the original virtual vector with an equal-amplitude and in-phase redundant virtual vector to adjust the duration of the midpoint current loop and achieve capacitor midpoint voltage balance;

[0012] S4: Adjust the output order of each basic vector through an interleaved PWM modulation method to suppress the number of switchings during control period alternation.

[0013] Further, the construction of the virtual vector set in S1 specifically includes:

[0014] Synthesize a virtual vector in the double αβ coordinate system so that the periodic average voltage vectors in the α1β1 and α2β2 coordinate systems are in phase and equal in amplitude, cancel the harmonic voltage components in the xy sub-plane, and generate a set of 12 virtual vectors through the synthesis path shown in Table 1.

[0015] Furthermore, the construction of the redundant virtual vector includes:

[0016] Generate a redundant virtual vector by adjusting the synthesis ratio of the native medium vector and the adjacent large vector, where the balance factor w is used to control the duration of the midpoint current loop, and the value range of the balance factor w is 0 < w < 1, and it is calculated in real time through the following formula:

[0017]

[0018] In the formula, U dc is the bus voltage, u c2 is the lower bridge arm capacitor voltage, i o is the instantaneous midpoint current, T s is the control period.

[0019] Furthermore, the interleaved PWM modulation of S4 specifically includes:

[0020] When the control period alternates, define the switching times cost function g s , traverse the alternative switching states to select the vector output order with the fewest switching times during the period alternation, and determine the timing of each basic vector through the interleaved PWM output schematic diagram.

[0021] Furthermore, the method also includes a dynamic vector adjustment step:

[0022] Predict the alternative vectors in the next period according to the optimal vector in the current period, and optimize the dq-axis current tracking performance through the cost function g dq .

[0023] A three-level dual three-phase PMSM system based on multi-objective decoupling optimization includes:

[0024] A virtual vector generation module for constructing a virtual vector set with low common-mode voltage and zero harmonic voltage components;

[0025] A deadbeat control module for calculating the balance factor w when the capacitor midpoint voltage fluctuation exceeds the limit and generating a redundant virtual vector;

[0026] An interleaved PWM modulation module for adjusting the output order of the basic vectors to suppress the number of switchings;

[0027] A dynamic optimization module for optimizing the dq-axis current tracking through the cost function g dq .

[0028] Furthermore, the virtual vector generation module is specifically configured as follows:

[0029] Synthesize virtual vectors in the double αβ coordinate system, and generate a set of 12 virtual vectors through the synthesis path shown in Table 1, where the common-mode voltage of each virtual vector is U dc / 12.

[0030] Furthermore, the deadbeat control module further includes:

[0031] A midpoint current calculation unit for calculating the instantaneous midpoint current i o in real time according to the switching state, and adjusting the proportion of the medium vector in the redundant virtual vectors through the balance factor w.

[0032] Furthermore, the interleaved PWM modulation module is specifically configured as follows:

[0033] According to the alternative switching state quantities during period alternation, select the output order with the fewest number of switchings through the switching times cost function g s and adjust the output order of each basic vector based on the preset interleaved timing.

[0034] Furthermore, the dynamic optimization module further includes:

[0035] A delay compensation unit for compensating the current prediction value based on the historical optimal vector to improve the dq-axis current tracking accuracy..

[0036] The beneficial effects of the present invention are as follows: Based on the constructed virtual vectors with low common-mode voltage and zero harmonic voltage characteristics, and adopting the interleaved pulse width modulation (PWM) modulation method, the present invention realizes the cost-function-free suppression of the common-mode voltage, harmonic current, and switching frequency. Through the designed redundant virtual vectors combined with the deadbeat controller, the active capacitor midpoint potential balance and the decoupling of the flux linkage and torque control are realized. Finally, the cost function focuses on the optimal vector search for dq current tracking, thereby realizing the decoupled optimal control of all sub-goals. The proposed method significantly improves the dq-axis current tracking accuracy and dynamic and steady-state performance through the independent decoupled control of sub-goals, and has good multi-objective optimization ability. Among them, the sub-common-mode voltage is reduced to U dc / 6, the total common-mode voltage is reduced to U dc / 12, the phase current THD is lower than 13%, and the highest switching frequency is only 4.3 kHz.

[0037] Other advantages, objects, and features of the present invention will be set forth in part in the description which follows, and in part will be obvious to those having ordinary skill in the art upon examination of the following, or may be learned from practice of the present invention. The objects and other advantages of the present invention may be realized and obtained by the means of the instrumentalities and combinations particularly pointed out hereinafter. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] In order to make the objectives, technical solutions, and advantages of the present invention more clear, the present invention will be described in detail below with reference to the accompanying drawings, where:

[0039] Figure 1 is the basic voltage vector in the double αβ coordinate system; (a) is the space voltage vector in the α1β1 coordinate system; (b) is the space voltage vector in the α2β2 coordinate system;

[0040] Figure 2 is the alternative virtual voltage vector;

[0041] Figure 3 is the schematic diagram of the redundant vector synthesis path;

[0042] Figure 4 is the alternative switch state quantity at the periodic alternating moment;

[0043] Figure 5 is the schematic diagram of the interleaved PWM output; (a) is Case 1, the alternating process of the PWM sequence ending with V 16 and V0; (b) is Case 2, the alternating process of the PWM sequence ending with V 15 and V 26 ;

[0044] Figure 6 are the steady-state experimental waveforms when operating at 200 rpm and 500 rpm respectively with a load of 4 Nm;

[0045] Figure 7 is the dynamic response waveform during the process of the load stepping from no load to the rated 4 Nm load at 500 rpm;

[0046] Figure 8 is the dynamic response waveform during the process of stepping from stationary to 500 rpm with a load of 4 Nm. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0047] The following specific examples illustrate the implementation manners of the present invention. Those skilled in the art can easily understand other advantages and effects of the present invention from the content disclosed in this specification. The present invention can also be implemented or applied through other different specific implementation manners. Various details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of the present invention. It should be noted that the drawings provided in the following embodiments only schematically illustrate the basic concept of the present invention. Without conflict, the following embodiments and the features in the embodiments can be combined with each other.

[0048] Among them, the drawings are only for illustrative purposes, showing only schematic diagrams, rather than physical diagrams, and should not be construed as a limitation on the present invention; in order to better illustrate the embodiments of the present invention, some components in the drawings will be omitted, enlarged or reduced, which does not represent the size of the actual product; for those skilled in the art, it is understandable that some well-known structures and their descriptions in the drawings may be omitted.

[0049] In the drawings of the embodiments of the present invention, the same or similar reference numerals correspond to the same or similar components; in the description of the present invention, it should be understood that if there are terms such as "upper", "lower", "left", "right", "front", "rear", etc. indicating the orientation or positional relationship, they are based on the orientation or positional relationship shown in the drawings. It is only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation. Therefore, the terms describing the positional relationship in the drawings are only for illustrative purposes and should not be construed as a limitation on the present invention. For those of ordinary skill in the art, the specific meanings of the above terms can be understood according to specific circumstances.

[0050] The proposed method divides the control objectives into four parts:

[0051] 1) Harmonic, common-mode voltage and switching frequency suppression within the control period;

[0052] 2) dq-axis current tracking optimization;

[0053] 3) Capacitor midpoint potential balance;

[0054] 4) Switching frequency suppression during the alternation of control periods.

[0055] For each objective, the following optimization methods are adopted:

[0056] A virtual vector set with zero harmonic voltage and low common-mode voltage characteristics, an independent cost function g dq , a redundant virtual vector set that is in phase and equal in amplitude with the virtual vector but whose influence on capacitor midpoint potential balance is adjustable, and an interleaved PWM modulation method.

[0057] The specific process is as follows: First, delay compensation is performed on the current based on the historical optimal vector (the optimal vector of cycle k), and then the cost function g is used to dq traverse the optimal virtual vectors. When it is detected that the balance fluctuation of the midpoint potential of the capacitor exceeds the threshold of 5% of the bus voltage (U dc ), the deadbeat controller is started to calculate the balance factor w, and the native virtual vector is replaced with an equal-amplitude and in-phase redundant virtual vector. Finally, the output order of each basic vector is adjusted by the interleaved PWM modulation method.

[0058] 1. Virtual Vector Construction

[0059] In the traditional scheme, virtual vector synthesis is usually carried out in the vector space decoupled coordinate system. In this coordinate system, the phase current and voltage are mapped to three sub-planes: αβ, xy, and o1o2. Among them, αβ is the electromechanical energy conversion plane, xy is the harmonic mapping sub-plane, and o1o2 is uncontrollable. For simplicity of analysis, according to the three-phase Clarke transformation, the variables (specifically referring to the phase voltage here) in the six-phase stator coordinate system are mapped into two independent αβ coordinate systems (αβ1 plane and αβ2 plane). The calculation equations are as follows:

[0060]

[0061] In the formula, S x (x = a~f) represents three switching states P, O, and N output by the six-phase bridge arm, and are represented by S x = 1, 0, -1 respectively. According to (1), the basic space voltage vectors in the double αβ coordinate system shown in Figure 1 are obtained. Figure 1 is the basic voltage vector in the double αβ coordinate system; (a) is the space voltage vector in the α1β1 coordinate system; (b) is the space voltage vector in the α2β2 coordinate system.

[0062] The conversion relationship between the vector space decoupled coordinate system and the double αβ coordinate system is as follows

[0063]

[0064] It can be seen from formula (2) that the periodic average voltage vectors of the synthesized virtual vectors in the α1β1 and α2β2 coordinate systems should be in-phase and equal-amplitude vectors so that the voltage components in the xy sub-plane can be cancelled out, thereby meeting the demand for the 0 harmonic voltage component in the xy sub-plane. And, in order to minimize the number of switching actions caused by the switching of the basic vector states, as few basic vectors as possible should be used to synthesize the virtual vector. Finally, the virtual vector synthesis path in Table 1 and Figure 2 the alternative virtual voltage vectors in

[0065] Table 1 Synthesis Path of Virtual Vector Set

[0066] vector synthesis path common mode voltage vector synthesis path common mode voltage <![CDATA[VV1]]> <![CDATA[aV 15 +V 26 > <![CDATA[0,U dc / 12]]> <![CDATA[VV2]]> <![CDATA[V 16 +aV 15 > <![CDATA[0,U dc / 12]]> <![CDATA[VV3]]> <![CDATA[aV 17 +V 16 > <![CDATA[0,U dc / 12]]> <![CDATA[VV4]]> <![CDATA[V 18 +aV 17 > <![CDATA[0,U dc / 12]]> <![CDATA[VV5]]> <![CDATA[aV 19 +V 18 > <![CDATA[0,U dc / 12]]> <![CDATA[VV6]]> <![CDATA[V 20 +aV 19 > <![CDATA[0,U dc / 12]]> <![CDATA[VV7]]> <![CDATA[aV 21 +V 20 > <![CDATA[0,U dc / 12]]> <![CDATA[VV8]]> <![CDATA[V 22 +aV 21 > <![CDATA[0,U dc / 12]]> <![CDATA[VV9]]> <![CDATA[aV 23 +V 22 > <![CDATA[0,U dc / 12]]> <![CDATA[VV 10 > <![CDATA[V 24 +aV 23 > <![CDATA[0,U dc / 12]]> <![CDATA[VV 11 > <![CDATA[aV 25 +V 24 > <![CDATA[0,U dc / 12]]> <![CDATA[VV 12 > <![CDATA[V 26 +aV 25 > <![CDATA[0,U dc / 12]]>

[0067] Note: a = 0.866, VV0 is the zero vector, and the switch state set S = [000000].

[0068] 2. Construction of Redundant Virtual Vectors

[0069] By constructing redundant middle vectors that are in phase and equal in amplitude to the middle vector, but with a controllable duration of the midpoint current loop, independent regulation of the capacitor midpoint potential balance can be achieved. Specifically, the original middle vector V M and two adjacent large vectors V L can be further synthesized into redundant virtual vectors. By adjusting the proportion of the original middle vector in the redundant virtual vector, the duration of the midpoint current loop can be indirectly controlled. Without affecting torque and flux linkage control, active regulation of the capacitor midpoint potential balance can be achieved, and then decoupled independent control of the capacitor midpoint potential balance can be realized. Taking V 16 in the α1β1 coordinate system as an example, Figure 3 shows the synthesis path of the redundant virtual middle vector V′ 16 .

[0070] The redundant middle vector V' M is composed of two adjacent large vectors and a part of the middle vector, and the synthesis ratio is as follows

[0071] V' M = w·V M + 0.5(1 - w)·(V L1 + V L2 ) (3)

[0072] In the formula, V L1 and V L2 represent two adjacent large vectors. w (0 < w < 1) is the proportion of the basic middle vector V M in the redundant middle vector. By adjusting w, the capacitor midpoint potential balance can be regulated, so w is named the capacitor midpoint potential balance factor. The instantaneous value of the midpoint current i o can be calculated by the instantaneous switch state acting on the motor:

[0073]

[0074] Taking the lower bridge arm capacitor voltage u c2 = 0.5U dc as the tracking target of the deadbeat controller, the value of the balance factor w at this time can be calculated as:

[0075]

[0076] The value range of w is 0 < w < 1. If it exceeds this range, the upper and lower limits of the value range are taken respectively. The value of w can be calculated in real time by the deadbeat controller to make u c2 = 0.5U dc , thus achieving the balance of the midpoint potential of the capacitor.

[0077] 3. Interleaved PWM Modulation

[0078] The switching actions occur during the vector switching process inside the control period and at the alternation of control periods. The former is suppressed by the virtual vector, while the latter requires controlling the output sequence of the vectors cycle by cycle. For this reason, an interleaved PWM modulation method is proposed.

[0079] At the starting moment of cycle k + 1, there are multiple choices for the switching state . Although the average voltages output in cycle k + 1 are the same for different choices, it will affect the number of switching actions of the switching tubes during the alternation between cycle k and cycle k + 1. Therefore, a cost function g s is defined to evaluate the number of switching actions during the cycle switching process.

[0080]

[0081] Suppose that at the end of cycle k, there are Figure 4 two groups of switching states which are [V 16 V0] and [V 15 V 26 respectively, and their alternative vector states are as follows

[0082] Through g s , the vector output sequence with the minimum number of switchings during the alternation between cycle k + 1 and cycle k can be traversed. Figure 4 The PWM sequence order of the optimal vector traversed from the vectors in Figure 5 is shown as follows. (a) is Case 1, the alternation process of the PWM sequence when ending with V 16 and V0; (b) is Case 2, the alternation process of the PWM sequence when ending with V 15 and V 26 .

[0083] Verification experiments are carried out on the T-type three-level dual-three-phase motor pair test platform under the rated steady-state and dynamic conditions with the rated load (4 Nm). Finally, the experimental effect diagrams under various working conditions of Figures 6 to 8 are obtained, indicating that the proposed method can effectively achieve multi-objective optimization.

[0084] It can be seen from the steady-state and dynamic experimental results that the multi-objective decoupling priority model predictive control method proposed in this paper has good harmonic suppression effect, and on the premise of satisfying the dq command following, it can effectively reduce the full-speed range switching frequency to below 4.3 kHz and the CMV to U dc / 12.

[0085] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical solutions of the present invention can be modified or equivalently replaced without departing from the purpose and scope of the present technical solution, and they should all be covered by the scope of the claims of the present invention.

Claims

1. A FCS-MPC method for a three-level dual-three-phase PMSM system based on multi-objective decoupling optimization, characterized in that: It includes the following steps: S1: Construct a set of virtual vectors with low common-mode voltage and zero harmonic voltage components to suppress harmonic current, common-mode voltage, and the switching frequency within the control period; S2: According to the set of virtual vectors, traverse the optimal virtual vector through a cost function, and perform delay compensation on the current based on the historical optimal vector; S3: Detect the fluctuation of the capacitor midpoint voltage. If it exceeds the preset threshold, start a deadbeat controller to calculate the balance factor w, and replace the original virtual vector with a redundant virtual vector of equal amplitude and in-phase to adjust the duration of the midpoint current loop and achieve capacitor midpoint voltage balance; S4: Adjust the output order of each basic vector through an interleaved PWM modulation method to suppress the number of switches during the alternation of control periods.

2. The FCS-MPC method for a three-level dual-three-phase PMSM system based on multi-objective decoupling optimization according to claim 1, wherein: The construction of the virtual vector set in S1 specifically includes: Synthesize virtual vectors in the double αβ coordinate system so that the periodic average voltage vectors in the α1β1 and α2β2 coordinate systems are in-phase and of equal amplitude to cancel the harmonic voltage components in the xy sub-plane, and generate a set of 12 virtual vectors through the synthesis path shown in Table 1.

3. The FCS-MPC method for a three-level dual-three-phase PMSM system based on multi-objective decoupling optimization according to claim 1, characterized in that: The construction of the redundant virtual vector includes: Generate a redundant virtual vector by adjusting the synthesis ratio of the original medium vector and the adjacent large vector, where the balance factor w is used to control the duration of the midpoint current loop, and the value range of the balance factor w is 0 < w < 1, and it is calculated in real time through the following formula: Where, U dc is the bus voltage, u c2 is the capacitor voltage of the lower arm, i o is the instantaneous midpoint current, T s is the control period.

4. The FCS-MPC method for a three-level dual-three-phase PMSM system based on multi-objective decoupling optimization according to claim 1, characterized in that: The interleaved PWM modulation in S4 specifically includes: When the control cycles alternate, define the switching times cost function g s , traverse the alternative switching states to select the vector output sequence with the fewest switching times during the cycle alternation, and determine the timing of each basic vector through the interleaved PWM output schematic diagram.

5. The FCS-MPC method for a three-level dual-three-phase PMSM system based on multi-objective decoupling optimization according to claim 1, characterized in that: The method further includes a dynamic vector adjustment step: Predict the candidate vectors for the next period based on the optimal vector of the current period, and optimize the dq-axis current tracking performance through the cost function g dq Optimize the dq-axis current tracking performance.

6. A three-level dual three-phase PMSM system based on multi-objective decoupling optimization, characterized in that: It includes: A virtual vector generation module for constructing a set of virtual vectors with low common-mode voltage and zero harmonic voltage components; A deadbeat control module for calculating the balance factor w when the capacitor midpoint voltage fluctuation exceeds the limit and generating a redundant virtual vector; An interleaved PWM modulation module for adjusting the output order of the basic vectors to suppress the number of switches; A dynamic optimization module for optimizing the dq-axis current tracking through a cost function g dq Optimize the dq-axis current tracking.

7. The three-level dual-three-phase PMSM system based on multi-objective decoupling optimization according to claim 6, characterized in that: The virtual vector generation module is specifically configured as: Synthesize virtual vectors in the double αβ coordinate system and generate a set of 12 virtual vectors through the synthesis path shown in Table 1, where the common-mode voltage of each virtual vector is U dc / 12.

8. The three-level dual-three-phase PMSM system based on multi-objective decoupling optimization according to claim 6, wherein: The deadbeat control module further includes: Midpoint current calculation unit, which is used to calculate the instantaneous midpoint current i in real time according to the switch states o , and adjust the proportion of the medium vector in the redundant virtual vectors through the balance factor w.

9. The three-level dual-three-phase PMSM system based on multi-objective decoupling optimization according to claim 6, characterized in that: The interleaved PWM modulation module is specifically configured as: Based on the alternative switch state quantities during cycle alternation, select the output sequence with the fewest number of switchings through the switching times cost function g s Select the output sequence with the fewest number of switchings and adjust the output sequences of the basic vectors based on a preset interleaved timing sequence.

10. The three-level dual-three-phase PMSM system based on multi-objective decoupling optimization according to claim 6, characterized in that: The dynamic optimization module further includes: A delay compensation unit for compensating the current prediction value based on the historical optimal vector to improve the dq-axis current tracking accuracy.

Images