Reconfiguration fault-tolerant method for dual permanent magnet synchronous motor based on nine-switch converter

Through nine-switch converter and model prediction control, the problem of the large number of current sensors in the dual permanent magnet synchronous motor drive system is solved, and efficient current reconstruction and fault diagnosis is achieved, reducing costs and improving power density and reliability.

CN118677332BActive Publication Date: 2024-11-08HEBEI WANGXIN GOVERNMENT SOFTWARE CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411154434.7
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-08-22
Publication Date
2024-11-08
Estimated Expiration
2044-08-22

AI Technical Summary

Technical Problem

The existing dual permanent magnet synchronous motor drive system has a large number of power switching devices and current sensors, resulting in high cost and low power density. The existing single current sensor current reconstruction method is only applicable to three-phase half-bridge circuit topology and cannot meet the dual The motor independently controls the requirements.

Method used

The fault tolerance method of dual permanent magnet synchronous motor based on nine-switch converter is adopted. The three-phase current is reconstructed through nine-switch converter, and a single current sensor is used for fault diagnosis and fault tolerance control, and the current-free sensor control is realized in combination with model prediction control. .

Benefits of technology

It greatly reduces the number of power devices and current sensors, improves system reliability and power density, reduces system costs, and realizes high-performance driving with independent control of dual motors.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118677332B_ABST
    Figure CN118677332B_ABST
Patent Text Reader

Abstract

The present invention belongs to the field of motor system and control technology, and specifically provides a fault-tolerant reconstruction method for a dual permanent magnet synchronous motor based on a nine-switch converter. The phase shift method, voltage vector insertion method and approximation method are used to reconstruct the current in the unobservable area of ​​sector switching, the unobservable area of ​​low modulation and the unobservable area of ​​high modulation respectively. And the fault diagnosis of the current sensor of the nine-switch converter is performed based on the stator coordinate transformation. In view of the problem of three-phase current reconstruction, a new voltage vector table is established, and a current observer is constructed based on the switching function and the mathematical model of the motor. In order to improve the dynamic performance and steady-state performance of the nine-switch converter dual-motor drive system, an improved finite set model predictive control is used to realize independent control of the dual motors. In this way, the cost is reduced, the power density is increased, and the reliability is improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention belongs to the technical field of motor systems and control, and specifically provides a fault-tolerant reconstruction method for a dual permanent magnet synchronous motor based on a nine-switch converter. Background Art

[0002] Permanent magnet synchronous motors are widely used in high-precision control applications such as industrial robots and CNC machine tools due to their outstanding features such as high power density, high reliability and high efficiency. At present, the most commonly used driving method for permanent magnet synchronous motors is the vector control method based on a three-phase half-bridge circuit. In the field of dual permanent magnet synchronous motor control, two three-phase half-bridge circuits are required to drive the motors, with a large number of 12 power devices and 6 current sensors, high cost and low power density.

[0003] As people's demands for product portability, operability, economy, reliability and energy saving increase, reducing costs, reducing volume and loss have become the hot research directions of engineering technology in the world today. Similarly, for the converter of the motor drive system, low-cost, high-reliability, low-switching devices have attracted the attention of experts and scholars in power electronics, and various new inverter topologies, various single current sensor current reconstruction methods and current sensorless control methods have also emerged.

[0004] Among the existing low-switch converter technology solutions, the five-bridge-arm converter has the largest number of power devices, and the four-bridge-arm converter has the least number of power devices, but has two more capacitors and a neutral point drift problem. Moreover, these circuits require a large number of current sensors when driving motors. The current single current sensor current reconstruction method and fault diagnosis and fault-tolerant control method are only applicable to three-phase half-bridge circuit topology, and two motors require two current sensors. It is necessary to develop a dual-motor independent control method with fewer power switching devices and current sensors to reduce costs and improve reliability. Summary of the invention

[0005] In order to overcome the above defects, the present invention provides a low-cost dual permanent magnet synchronous motor reconfiguration fault-tolerant method based on nine-switch converter.

[0006] In a first aspect, the present invention provides a fault-tolerant reconstruction method for a dual permanent magnet synchronous motor based on a nine-switch converter, comprising:

[0007] The three-phase current of the dual permanent magnet synchronous motor is reconstructed based on a nine-switch converter; wherein the nine-switch converter includes a first switch element, a second switch element, a third switch element, a fourth switch element, a fifth switch element, a sixth switch element, a seventh switch element, an eighth switch element, a ninth switch element, a resistor, a capacitor and a current sensor;

[0008] Sampling the current of the synchronous motor based on the current sensor, and determining whether the current sensor fails;

[0009] If a fault occurs, a current observer is used to perform fault-tolerant control based on the current sensor.

[0010] Furthermore, the nine-switch converter comprises:

[0011] The input end of the first switch element is connected to the first end of the capacitor, the output end of the first switch element is connected to the input end of the second switch element, the output end of the second switch element is connected to the input end of the third switch element, the control end of the first switch element receives a control signal to change the working state of the first switch element, the control end of the second switch element receives a control signal to change the working state of the second switch element, and the control end of the third switch element receives a control signal to change the working state of the third switch element;

[0012] The input end of the fourth switch element is connected to the first end of the capacitor, the output end of the fourth switch element is connected to the input end of the fifth switch element, the output end of the fifth switch element is connected to the input end of the sixth switch element, the control end of the fourth switch element receives a control signal to change the working state of the fourth switch element, the control end of the fifth switch element receives a control signal to change the working state of the fifth switch element, and the control end of the sixth switch element receives a control signal to change the working state of the sixth switch element;

[0013] The input end of the seventh switch element is connected to the first end of the capacitor, the output end of the seventh switch element is connected to the input end of the eighth switch element, the output end of the ninth switch element is connected to the input end of the sixth switch element, the control end of the seventh switch element receives a control signal to change the working state of the seventh switch element, the control end of the eighth switch element receives a control signal to change the working state of the eighth switch element, and the control end of the ninth switch element receives a control signal to change the working state of the ninth switch element;

[0014] The output end of the third switch element, the output end of the sixth switch element and the output end of the ninth switch element are all connected to the first end of the resistor, and the second end of the resistor is connected to the second end of the capacitor;

[0015] The first end and the second end of the resistor are respectively connected to the current sensor;

[0016] The first end of the capacitor is used to be connected to the positive electrode of the power supply, and the second end of the capacitor is used to be connected to the negative electrode of the power supply;

[0017] The output end of the first switching element is used to connect to the first phase line of the first motor, the output end of the fourth switching element is used to connect to the second phase line of the first motor, the output end of the seventh switching element is used to connect to the third phase line of the first motor, the output end of the second switching element is used to connect to the first phase line of the second motor, the output end of the fifth switching element is used to connect to the second phase line of the second motor, and the output end of the eighth switching element is used to connect to the third phase line of the second motor.

[0018] Furthermore, the reconstructing the three-phase current of the dual permanent magnet synchronous motor based on the nine-switch converter includes:

[0019] In the observable area and the non-observable area, the current sensor based on the nine-switch converter collects the three-phase current of the dual permanent magnet synchronous motor to complete the reconstruction of all phase currents. The collection method is to sample four times per cycle, and each collection obtains the two-phase currents of the two motors, for a total of four currents.

[0020] Furthermore, in the non-observation area, collecting the three-phase current of the dual permanent magnet synchronous motor based on the current sensor of the nine-switch converter includes:

[0021] When the voltage vector is located in the non-observation area of ​​sector switching, the action time of a single voltage vector is adjusted;

[0022] When the voltage vector is located in the low-modulation non-observation area, the action time of the two voltage vectors is adjusted. Only one voltage vector is adjusted in each cycle, and the action time of the other voltage vector adopts the action time of the previous cycle. When the voltage vector is located in the high-modulation non-observation area, the voltage vector is adjusted based on the approximate processing method.

[0023] Furthermore, when the voltage vector is located in the sector switching non-observation area, adjusting the action time of a single voltage vector includes:

[0024] The voltage vector action time of the first motor is T1, and the voltage vector action time of the second motor is T2. min is the minimum sampling time, T min =T d +T set +T adc ; T d Indicates dead time; T set Indicates the bus current establishment time; T adc Indicates the time required for sampling.

[0025] When T 1 ≥2T min And T min <T 2 <2T minWhen the voltage is equal to the input voltage, the action time of a single voltage vector is adjusted based on the phase shift method;

[0026] When T 1 ≥2T min And T 2 <2T min When the voltage vector is inserted, the action time of a single voltage vector is adjusted.

[0027] Further, sampling the current of the synchronous motor based on the current sensor to determine whether the current sensor fails includes:

[0028] The current sensors are used to sample the current of each phase of the two motors to obtain the six-phase currents of the two motors;

[0029] Obtaining a current residual of the sampled current;

[0030] Based on the difference between the current residual and a preset threshold, it is determined whether the current sensor fails.

[0031] Further, judging whether the current sensor fails based on the difference between the current residual and a preset threshold value includes:

[0032] If the current residual of any phase current is higher than a preset threshold, it is determined that the current sensor is faulty.

[0033] Further, the fault-tolerant control based on the current sensor includes:

[0034] Construct a current observer;

[0035] Obtaining three-phase current values ​​of two motors based on the current observer;

[0036] According to the obtained three-phase current values, fault-tolerant control is performed based on the improved model predictive control.

[0037] Furthermore, the improved model predictive control includes:

[0038] Obtaining voltage vector reference angles of the first motor and the second motor;

[0039] Establishing a selection strategy according to the voltage vector reference angle, selecting 5 voltage vectors for traversal from 15 voltage vectors;

[0040] Based on the objective function, five voltage vectors are traversed, and the switch state corresponding to the voltage vector with the minimum value of the objective function is selected as the switch state of the next cycle.

[0041] Furthermore, the voltage vector reference angle θ v * :

[0042] ;

[0043] In the formula, is the power angle, ; According to the speed loop feedback i * q We can get ψ * d and ψ * q θ e is the stator electrical angle;

[0044] The objective function:

[0045] ;

[0046] Among them, the objective function of the current prediction of the two motors is obtained by subtracting the dq-axis current at time k+1 from the q-axis current reference value and the d-axis reference current obtained by the two speed loops:

[0047]

[0048]

[0049] λ is the weight coefficient of the two motors, and μ is the weight coefficient of the loss function.

[0050] The above one or more technical solutions of the present invention have at least one or more of the following beneficial effects: In the implementation of the technical solution of the present invention, the phase current reconstruction method of the dual permanent magnet synchronous motor single current sensor based on the nine-switch converter greatly reduces the number of power devices and current sensors. The current sensor fault diagnosis and fault-tolerant control based on the nine-switch model predictive control improves the reliability of the dual motor drive system based on the nine-switch converter. Moreover, the three-phase current fault-tolerant control based on the model predictive control is equivalent to the current sensor-free control, which further reduces the system cost, improves the power density, and improves the overall performance of the dual motor control. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] The disclosure of the present invention will become more easily understood with reference to the accompanying drawings. It is easy for those skilled in the art to understand that these drawings are only for illustrative purposes and are not intended to limit the scope of protection of the present invention. In addition, similar numbers in the figures are used to represent similar components, among which:

[0052] Figure 1 It is a flowchart of main steps of a method for reconfiguring a double permanent magnet synchronous motor with a nine-switch converter to tolerate faults according to an embodiment of the present invention;

[0053] Figure 2is a main structural schematic diagram of a nine-switch converter according to an embodiment of the present invention;

[0054] Figure 3 is a vector sampling timing diagram of a phase shift method according to an embodiment of the present invention;

[0055] Figure 4 is a schematic diagram of a voltage vector insertion method according to an embodiment of the present invention;

[0056] Figure 5 is a voltage vector insertion normal vector sampling timing diagram according to an embodiment of the present invention;

[0057] Figure 6 is a schematic diagram of a case where the α coordinate axis in the αβ coordinate system coincides with the directions of the three-phase voltages respectively according to an embodiment of the present invention;

[0058] Figure 7 is a schematic diagram of the coincidence of the α axis and a in the αβ coordinate system according to an embodiment of the present invention;

[0059] Figure 8 is a schematic diagram of the coincidence of the α axis and c in the αβ coordinate system according to an embodiment of the present invention;

[0060] Fig. 9 is a nine-switch dual-motor stator current fault diagnosis diagram based on current residual according to an embodiment of the present invention;

[0061] Fig.10 is a schematic diagram of a dual-motor dq-axis model of a switching converter according to an embodiment of the present invention;

[0062] Fig.11 is a schematic diagram of a current observer model according to an embodiment of the present invention;

[0063] Fig.12 is a space vector diagram of a permanent magnet synchronous motor according to an embodiment of the present invention;

[0064] Fig.13 is a schematic diagram of a voltage vector screening process according to an embodiment of the present invention;

[0065] Fig.14 is the reconstructed current and the actual current of the A phase of the first motor when the load is 1 N·m according to an embodiment of the present invention;

[0066] Fig.15 is the reconstructed current and the actual current of the A phase of the first motor when the load is 5 N·m according to an embodiment of the present invention;

[0067] Fig.16is the A-phase reconstructed current and the actual current of the second motor when the load is 1 N·m according to an embodiment of the present invention;

[0068] Fig.17 is the reconstructed current and actual current of the A phase of the second motor when the load is 5 N·m according to an embodiment of the present invention;

[0069] Fig.18 is a schematic diagram of the main steps of executing a single resistor sampling program according to an embodiment of the present invention;

[0070] Fig.19 is a schematic diagram of speed waveforms of dual motors under fault-tolerant control according to an embodiment of the present invention;

[0071] Fig. 20 is a schematic diagram of reconstructed current and sampled current of phase A of a motor according to an embodiment of the present invention;

[0072] Fig.21 is a schematic diagram of observed current and sampled current of the motor q axis according to an embodiment of the present invention;

[0073] Fig. 22 The waveform of the motor speed in a dynamic experiment with continuous step-change speed according to an embodiment of the present invention;

[0074] Fig.23 is a current waveform of phase a of the motor in a dynamic experiment with continuous step change of speed according to an embodiment of the present invention;

[0075] Fig.24 is a waveform of the motor speed in a load step change experiment according to an embodiment of the present invention;

[0076] Fig.25 1 is a current waveform of phase a of the upper motor in a load step change experiment according to an embodiment of the present invention. DETAILED DESCRIPTION

[0077] Some embodiments of the present invention are described below with reference to the accompanying drawings. It should be understood by those skilled in the art that these embodiments are only used to explain the technical principles of the present invention and are not intended to limit the protection scope of the present invention.

[0078] Reference Figure 1 The present invention provides a double permanent magnet synchronous motor reconfiguration fault tolerance method based on a nine-switch converter, comprising:

[0079] S1, reconstructing the three-phase current of the dual permanent magnet synchronous motor based on a nine-switch converter; wherein the nine-switch converter includes a first switch element, a second switch element, a third switch element, a fourth switch element, a fifth switch element, a sixth switch element, a seventh switch element, an eighth switch element, a ninth switch element, a resistor, a capacitor and a current sensor.

[0080] S2, sampling the current of the synchronous motor based on the current sensor, and determining whether the current sensor fails.

[0081] S3: If a fault occurs, a current observer is used to perform fault-tolerant control based on the current sensor.

[0082] First, the nine-switch converter of the present invention is described.

[0083] Reference Figure 2 , the nine-switch converter includes:

[0084] The first switching element S AH The input end of the first switch element S is connected to the first end of the capacitor C. AH The output terminal of the second switching element S AM The input terminal of the second switching element S AM The output terminal of the third switch element S AL The input terminal is connected to the first switching element S AH The control terminal receives a control signal to change the first switch element S AH In the working state, the second switching element S AM The control terminal receives a control signal to change the second switch element S AM In the working state, the third switching element S AL The control terminal receives a control signal to change the third switch element S AL Working status;

[0085] The fourth switching element S BH The input end of the fourth switch element S is connected to the first end of the capacitor C1. BH The output terminal of the fifth switch element S BM The input terminal of the fifth switching element S BH The output terminal of the sixth switch element S BL The input terminal of the fourth switching element S BH The control terminal receives a control signal to change the fourth switch element S BH In the working state, the fifth switching element S BH The control terminal receives a control signal to change the fifth switch element SBH In the working state, the sixth switching element S BL The control terminal receives a control signal to change the sixth switch element S BL Working status;

[0086] The seventh switching element S CH The input end of the seventh switch element S is connected to the first end of the capacitor C1. CH The output terminal of the eighth switching element S CM The ninth switching element S CL The output terminal of the sixth switch element S BL The seventh switching element S CH The control terminal receives a control signal to change the seventh switch element S CH In the working state, the eighth switching element S CM The control terminal receives a control signal to change the eighth switch element S CM In the working state, the ninth switching element S CM The control terminal receives a control signal to change the ninth switch element S CM Working status;

[0087] The third switching element S AM The output terminal of the sixth switch element S BL The output terminal and the ninth switch element S CM The output ends of are connected to the first end of the resistor R1, and the second end of the resistor R1 is connected to the second end of the capacitor C1;

[0088] The first end and the second end of the resistor R1 are respectively connected to the current sensor L1;

[0089] Current sensor L1 is used to collect current i bus .

[0090] The first end of the capacitor C1 is used to be connected to the positive electrode of the power supply, and the second end of the capacitor C1 is used to be connected to the negative electrode of the power supply;

[0091] The first switching element S AH The output terminal is used to connect to the first phase line A of the first motor M1 H The fourth switching element S BH The output terminal is used to connect to the second phase line B of the first motor M1 H The seventh switching element S CH The output terminal is used to connect to the third phase line C of the first motor M1 H The second switching element S AMThe output terminal is used to connect to the first phase line A of the second motor M2 L The fifth switching element S BH The output terminal is used to connect to the second phase line B of the second motor M2 L The eighth switching element S CM The output terminal is used to connect to the second motor S CM The third phase line C L connect.

[0092] In one embodiment, the first to ninth switch elements are IGBT tubes. In the first to ninth switch elements, the input end is the collector, the output end is the emitter, and the control end is the gate.

[0093] In the dual-motor drive system based on the nine-switch converter of the present invention, two motors share a DC bus. This design enables the system to effectively reduce the number of hardware devices and reduce the overall cost.

[0094] A nine-switch converter is used to replace the traditional dual three-phase half-bridge circuit. The nine-switch converter has higher flexibility and control accuracy, and can better meet the needs of the dual-motor drive system.

[0095] The present invention adopts a single current sensor sampling method, that is, two motors share one current sensor. Compared with the traditional single current sensor sampling method, this method can reduce one more current sensor.

[0096] By reducing the number of current sensors, the present invention achieves the goal of lower cost and higher power density. In the traditional method, each motor needs a current sensor, while the present invention only needs one, thereby significantly reducing the system cost.

[0097] In the traditional dual-motor drive system, the number of power switch devices and current sensors is large, which leads to high cost and low power density. The present invention adopts a current sensor, which effectively reduces the number of these hardware devices.

[0098] The failure of the current sensor may lead to control failure and affect the normal operation of the system. The present invention reduces the probability of failure and improves the reliability of the system by reducing the number of current sensors.

[0099] The following describes a process of reconstructing the three-phase current of the dual permanent magnet synchronous motor based on a nine-switch converter. Specifically, it includes:

[0100] S11, in the observable area and the non-observable area, the current sensor based on the nine-switch converter collects the three-phase current of the dual permanent magnet synchronous motor to complete the reconstruction of all phase currents, wherein the collection method is to sample four times per cycle, and each collection obtains the two-phase currents of the two motors (the first motor and the second motor), for a total of four currents.

[0101] In one embodiment, the present invention uses single current sensor phase current reconstruction, that is, in the motor drive, a single current sensor is used to reconstruct the three-phase current. The key lies in the current reconstruction problem in the non-observation area.

[0102] The non-observation area is divided into a sector switching non-observation area, a low modulation non-observation area and a high modulation non-observation area. In the non-observation area, the current sensor of the nine-switch converter collects the three-phase current of the dual permanent magnet synchronous motor, including:

[0103] S111, in the non-observation area of ​​sector switching, only one voltage vector has a short action time, so two-phase current collection can be achieved by adjusting once. Therefore, when the voltage vector is located in the non-observation area of ​​sector switching, the action time of a single voltage vector is adjusted.

[0104] In one embodiment, when the voltage vector is located in the sector switching non-observation area, adjusting the action time of a single voltage vector includes:

[0105] The voltage vector action time of the first motor M1 is T1, and the voltage vector action time of the second motor M2 is T2. min is the minimum sampling time, T min =T d +T set +T adc ; T d Indicates dead time; T set Indicates the bus current establishment time; T adc Indicates the time required for sampling.

[0106] When T 1 ≥2T min And T min <T 2 <2T min When the voltage is equal to the input voltage, the action time of a single voltage vector is adjusted based on the phase shift method;

[0107] When T 1 ≥2T min And T 2 <2T min When the voltage vector is inserted, the action time of a single voltage vector is adjusted.

[0108] The above two situations are explained below respectively.

[0109] The non-observation area of ​​sector switching can be divided into the following two cases, see Table 1.

[0110] Table 1 Description of two situations

[0111] .

[0112] In Table 1, V1, V3, and V5 may completely represent the voltage vector representing each cycle of the first motor M1.

[0113] (1) When T 1 ≥2T min And T min <T 2 <2T min When, refer to Figure 3 , using the phase shift method.

[0114] (2) When T 1 ≥2T min And T 2 <2T min When the voltage vector is inserted, the voltage vector insertion method is used. The principle is as follows Figure 4 As shown. It can be seen that this method introduces three voltage vectors, in order not to affect the original reference voltage vector V ref1 The three voltage vectors are distributed in an equilateral triangle. Two of them are in the same or opposite direction as the V4 and V6 vectors, and these two vectors can be introduced by increasing or decreasing the size of V4 and V6. Only one of them is a new vector, which uses a portion of the time of the zero vector V1-3. It can be seen that after the adjustment, the voltage vector of each cycle remains unchanged, and the PWM control cycle remains unchanged.

[0115] Vector sampling time diagram Figure 5 shown.

[0116] S112, when the voltage vector is in the low modulation non-observation area, the motor speed is low, which means that the current changes very little, and the current is approximately unchanged in two adjacent cycles. In each cycle, only the action time of one voltage vector is adjusted, and the action time of the other voltage vector adopts the amount of the previous cycle.

[0117] In the low-modulation non-observation area, the action time of the two voltage vectors is relatively short, so the two time serial ports need to be adjusted. When the voltage vector is in the low-modulation non-observation area, the motor speed is low, which means that the current changes very little, and the current is approximately unchanged in two adjacent cycles. At this time, the voltage vector insertion method is used to sample a single current, and the other phase current remains unchanged. This ensures that in the half cycle corresponding to each motor, only the voltage vector of the voltage of one phase current needs to be adjusted, and the modulation wave of the power switch tube will not be adjusted too much, affecting the current waveform. Therefore, the voltage vector adjustment method at this time is consistent with the sector switching non-observation area solution.

[0118] Since the voltage vector insertion method proposed above, its basic idea is V 3 The action time T 2 No compensation for T min , not enough time T min -T 2 The time T of the zero vector 0 To supplement. The dwell time corresponding to the zero vector is reduced, so the previous voltage vector insertion method cannot be used to adjust the high modulation area, and a high modulation non-observation area will appear. The processing method for the high modulation non-observation area adopts an approximate processing method. Therefore, when the voltage vector is in the high modulation non-observation area, the voltage vector is adjusted based on the approximate processing method.

[0119] Next, S2 is described, wherein the sampling of the current of the synchronous motor based on the current sensor to determine whether the current sensor fails includes:

[0120] S21, sampling each phase current of the two motors by the current sensor to obtain six-phase currents of the two motors;

[0121] S22, obtaining a current residual of the sampled current;

[0122] S23, judging whether the current sensor fails based on the difference between the current residual and a preset threshold.

[0123] When the three-current sensor sampling method is adopted in the present invention, unlike the current sensors of the traditional three-phase half-bridge circuit, the current sensor of each phase of the nine-switch converter in this embodiment needs to complete the sampling of the current of the phase of the two motors.

[0124] The present invention is a current diagnosis method for a nine-switch converter dual-motor drive system based on coordinate transformation. Figure 6-Figure 8 Three coordinate systems in . Figure 6-Figure 8 The α axis in the αβ coordinate system corresponds to the direction of the three-phase voltage.

[0125] Schematic diagram of overlap. Figure 6 The α axis coincides with the a axis. Figure 7 The α axis coincides with the b axis. Figure 8 The α axis coincides with c.

[0126] i α1-3 、i β1-3 、i * α1-3 、i * β1-3 are the measured current values ​​and command values ​​of the α and β axes of the three coordinate axes respectively; i d1-3 、i q1-3 、i* d1-3 、i * q1-3 are the dq axis current measurement value and command value respectively. The three coordinate axes are the coordinate axes when α coincides with the three-phase voltage directions of a, b, and c respectively.

[0127] i α1-3 Indicates i α1 、i α2 and i α3 ,i α1-3 、i * α1-3 These are the three current measurement values ​​and three current reference values ​​of one of the motors obtained in step S21. The meanings of the remaining physical quantities can be deduced in the same way.

[0128] In S22, the current residual is assumed to be:

[0129] .

[0130] The judgment method in S23 can be used to judge the state of the current sensor according to the following Table 2:

[0131] Table 2 Current sensor status judgment

[0132]

[0133] In Table 2: ε 1-3 is the threshold of the current sensor, and the threshold is selected to be about 0.13i * q When it exceeds the threshold, it indicates that the current sensor is faulty. When it is less than the threshold, it indicates that the current sensor is normal. The following analysis is made for the threshold value:

[0134] 1) When the current sensor is normal, the current residual ε 1-3 Approximately zero.

[0135] 2) When the current sensor fails, the current residual is:

[0136]

[0137] The current residual δ in the above analysis 1-3 is the value under ideal conditions. After comprehensive consideration of possible error factors and multiple tests, the threshold value of about 0.13i*q is selected. The experimental results show that this threshold can effectively diagnose the fault of the current sensor.

[0138] In one embodiment, in step 23, judging whether the current sensor fails based on the difference between the current residual and a preset threshold value includes:

[0139] If the current residual of any phase current is higher than a preset threshold, it is determined that the current sensor is faulty.

[0140] In the sampling scheme of the single current sensor of the present invention, among the six-phase currents of the two three-phase motors, if the control program is correct, as long as there is a problem with one phase current, it means that there is a problem with the current sensor.

[0141] Reference Fig. 9 , Fig. 9 This is a nine-switch dual-motor stator current fault diagnosis diagram based on current residual, where the subscript H represents the first motor and the subscript L represents the second motor. Let the total fault flag f=x+y+z, then when f=0, all current sensors are normal. When f=1, it means that one current sensor is faulty. When f=2, it means that two current sensors are faulty. When f=3, it means that all current sensors are faulty. In the single current sensor dual-motor drive system based on nine-switch transformation, when f=3, the current sensor is faulty.

[0142] In one embodiment, in step S3, the performing fault-tolerant control based on the current sensor includes:

[0143] S31, constructing a current observer;

[0144] S32, obtaining three-phase current values ​​of two motors based on the current observer;

[0145] S33, performing fault-tolerant control based on the improved model predictive control according to the obtained three-phase current values.

[0146] The specific process in step S31 includes:

[0147] First, the large capacitor is divided into two capacitors of equal size and connected in series to the positive and negative poles of the power supply, and the space between the two capacitors is the grounding pole. Then, taking phase A as an example, when analyzing the switch model of the nine-switch converter, the upper output (the first motor M1) and the lower output (the first motor M2) are considered separately. When analyzing the upper output, the middle tube and the lower tube are considered as one switch tube for analysis, and when analyzing the upper output, the middle tube and the lower tube are considered as one switch tube for analysis.

[0148] The motor three-phase voltage equation based on the switching function of the nine-switch converter can be obtained:

[0149]

[0150] Where: u JY It's J Y (J=A, B, C; Y=H, L) Output voltage. The three-phase voltage equation is converted into a two-phase voltage equation through coordinate transformation:

[0151] { u d H = U dc ( S dH − S dM * S dL ) u q H = U dc [ S Q − 0 . 5 * ( S dM * S q + S Q * S dL ) ] u d L = U dc ( S dH * S dM − S dL ) u q L = U dc [ − S q − 0 . 5 * ( S dH * S Q + S dM * S q ) ]

[0152] According to the mathematical model of the two-phase synchronous coordinate system of the motor, the current observer formula of the dual motor control of the nine-switch converter is established:

[0153]

[0154] From equations (4) and (5), we can get the equivalent model of the two-phase dq axis of the nine-switch converter, as follows: Fig.10 shown.

[0155] Combining the switching function and the motor model, the current observer model can be obtained by substituting equation (5) into equation (4): Fig.11 The current sensor can obtain the direct axis observation current i of the two-phase rotating coordinate system of the upper and lower outputs by collecting the DC bus voltage and speed information. dj and the quadrature axis observation current i qj Since the DC bus voltage is a constant value in general, the observer can effectively observe the stator current of the motor.

[0156] The specific process of step S32 includes:

[0157] When the current sensor fails, the observed current obtained by the current observer is used to implement motor control.

[0158] The specific process of step S33 includes:

[0159] In one embodiment, in step S33, the improved model predictive control includes:

[0160] S331, obtaining voltage vector reference angles of the first motor and the second motor;

[0161] S332, establishing a selection strategy according to the voltage vector reference angle, and selecting 5 voltage vectors for traversal from 15 voltage vectors;

[0162] S333, traversing five voltage vectors based on the objective function, and selecting the switch state corresponding to the voltage vector with the minimum value of the objective function as the switch state of the next cycle.

[0163] The dynamic performance of the dual-motor drive system of the nine-switch converter based on the current observer is poor, and due to the inability to change the switching frequency, its loss is large and cannot be optimized. Since the finite set model predictive control has many advantages. This control method can be used in combination with the current observer. At this time, the system can realize the current sensorless model predictive control based on the nine-switch converter, that is, the fault-tolerant control scheme for three-current sensor multi-phase current fault and single current sensor fault. However, the finite set model predictive control based on the nine-switch sensor has more switch states than the six-switch converter, and the operation speed of a single cycle is slow, so the voltage vector screening method is used to improve the operation speed. At the same time, the space vector distance function is introduced to improve the objective function and reduce the switching loss.

[0164] In one embodiment, since the voltage vector of the traditional nine-switch converter has relatively more cycles, as shown in Table 3, the amount of calculation for each cycle is large, so it is necessary to further narrow the selection range of the voltage vector. If the electrical angle of the desired voltage vector of the upper and lower outputs in the αβ coordinate system is known, the selection range of the voltage vector can be narrowed. This method only needs to select between four valid voltage vectors and a zero vector at the boundary of the sector where the upper and lower output electrical angles are located. The key to this method is how to calculate the voltage vector reference angle of the upper and lower motors.

[0165] Table 3 Voltage vector table of nine-switch converter

[0166]

[0167] Fig.12 is the space vector diagram of the permanent magnet synchronous motor, and the angular frequency of the synchronous rotating coordinate system is ω e , the power angle (or torque angle) is δ, and the stator electrical angle is θ e When the stator resistance is neglected, the stator flux ψ s and the voltage vector V on the stator s The phase angle difference is π / 2.

[0168] Depend on Fig.12 It can be concluded that the angle θ between the voltage vector and the spatial reference axis A is v It can be expressed as:

[0169]

[0170] Where δ=arctan(ψ q / ψ d ); θe can be approximately regarded as a constant within a sampling period T. This paper applies the control scheme of i*d=0, and then according to the i*q of the speed loop feedback, ψ * d =ψ f and ψ* q =L q i * q , thus finding δ * =arctan(ψ * q / ψ * d ), modifying equation (6) can obtain the desired voltage vector angle θ v * :

[0171]

[0172] In formula (7), θ is the angle between the d-axis and the A-axis, that is, the rotor position angle. After obtaining the voltage vector reference angle, the corresponding nine-switch converter standby non-zero voltage vectors are selected according to the sectors where the two output voltage vector reference angles are located. The selection strategy is shown in Table 4. According to the four non-zero voltage vectors corresponding to the two outputs plus the zero vector, only five voltage vectors need to be traversed in one cycle. Compared with the original traversal of 15 voltage vectors, the number of current calculations and the number of comparisons of the cost function are reduced, which significantly reduces the amount of calculation.

[0173] Table 4 Voltage vector screening strategy for nine-switch converter

[0174]

[0175] The corresponding control flow chart is as follows Fig.13 As shown. The objective function of the current prediction of the two motors is obtained by subtracting the dq axis current at time k+1 from the q axis current reference value and the d axis reference current obtained by the two speed loops:

[0176] .

[0177] In order to characterize the switching loss of the nine-switch converter, the concept of space vector distance function (SVDF) is introduced, which is also called loss function in the present invention. k and V k+1 The SVDF between them is as follows:

[0178]

[0179] The above formula can be summarized as:

[0180]

[0181] In summary, the objective function of the improved nine-switch converter dual-motor is as follows:

[0182]

[0183] Among them, λ is the weight coefficient of the two motors. By adjusting λ, priority control of different motors can be achieved. When λ is increased, the second motor can obtain better control effect; when λ is reduced, the first motor can obtain better control effect. μ is the weight coefficient of the loss function. Under the same conditions, the larger the μ, the more obvious the loss reduction effect, but the corresponding current harmonic content will increase. Therefore, it is necessary to select a suitable loss weight coefficient so that the system can reduce the loss without affecting the current harmonic content too much.

[0184] Based on the above analysis, when the current sensor fails, the actual sampled current at time k in the improved model predictive control is replaced by the observed current obtained by the current observer, realizing multi-phase current fault-tolerant control. That is, at this time, the finite set model predictive control of the nine-switch converter and the current observer can be combined to establish a model predictive control system for the nine-switch converter dual motor without current sensor.

[0185] The present invention adopts a single current sensor current reconstruction method based on a nine-switch converter to drive dual motors. At this time, the dual motor drive system only needs nine power devices and one current sensor. A current sensor fault diagnosis and fault-tolerant control method based on this method is proposed. The phase shift method, voltage vector insertion method and approximation method are used to reconstruct the current in the unobservable area of ​​sector switching, the unobservable area of ​​low modulation and the unobservable area of ​​high modulation respectively. And the fault diagnosis of the current sensor of the nine-switch converter is carried out based on the stator coordinate transformation. In view of the problem of three-phase current reconstruction, a new voltage vector table is established, and a current observer is constructed based on the switching function and the mathematical model of the motor. In order to improve the dynamic performance and steady-state performance of the nine-switch converter dual motor drive system, an improved finite set model predictive control is used to realize independent control of the dual motors. In this way, the cost is reduced, the power density is increased, and the reliability is improved.

[0186] In order to verify the method of the present invention, a dual motor control experiment was carried out based on the single current sensor sampling method of the nine-switch converter of the present invention. The parameters and working environment of the two motors are the same. In the present invention, the upper motor is the first motor M1, and the lower motor is the second motor M2. The steady-state speed of the first motor M1 is set to 200r / min, and the speed of the second motor M2 is set to 400r / min. Then, the reconstructed current and the actual sampled current are observed when the load is 1N·m and the load is 5N·m, respectively. Figure 14-15 is the experimental result of the first motor phase current reconstruction, where Fig.14 and Fig.15They are the reconstructed current and actual current of phase A with a load of 1 N·m and a load of 5 N·m, respectively. Figure 16-17 The experimental results of the second motor phase current reconstruction are as follows: Fig.16 and Fig.17 They are the reconstructed current and actual current of phase A with a load of 1 N·m and a load of 5 N·m, respectively.

[0187] It can be seen from the experiment that the single current sensor sampling method can also realize independent control of two motors. The sampling current and reconstructed current obtained by the two motors under different loads have the same phase and amplitude, but different current harmonics. This proves the effectiveness of the current reconstruction method proposed in the present invention.

[0188] The present invention is based on a current sensor fault diagnosis and fault-tolerant control experiment of a nine-switch converter.

[0189] 1) Software program design process

[0190] The basic flow chart designed based on the basic principles of dual motor current sensor fault diagnosis and fault-tolerant control is as follows: Fig.18 shown.

[0191] First, the controller is powered on and the basic configuration is initialized. First, the fault diagnosis of the current sensor of the dual motor system is performed. If the current sensor is normal, the original three-current sensor sampling method is maintained (the circuit structure of the three-current sensor is the same as Figure 2 The difference is that the third switching element S AL , the sixth switching element S BL , the ninth switching element S CM The output ends of the two resistors are connected to a resistor, and each resistor is connected to a current sensor, so that S AL , S BL , S CM There are three current sensors connected to the three branches. If a current sensor fails, a fault-tolerant control method is used.

[0192] 2) Experimental results analysis

[0193] (1) Current sensor fault diagnosis

[0194] After the motor is started, a load of 4 N·m is added. First, the current sensor of phase A fails at 5s; the sensor of phase B fails 5s after phase A; and the sensor of phase C fails 5s after phase B. Since there are many waveforms to be sampled, an oscilloscope is used in conjunction with a virtual serial port to observe the experimental results.

[0195] (2) Current sensor fault-tolerant control

[0196] When the program determines that two or more current sensors are in a faulty state, the system will switch to fault-tolerant control mode to avoid the motor losing control. The current obtained by the current observer is used as the required feedback signal to achieve model prediction of current sensorless control. The speeds of the upper and lower motors are set to 400r / min and 200r / min respectively. First, the experiment is carried out under steady-state conditions. Fig.19 Represents the speed waveform of the dual motors under fault-tolerant control; Fig. 20 Represents the reconstructed current and sampled current of phase A of the upper motor; Fig.21 Represents the observed current and sampled current of the q-axis of the upper motor. Fig.19 It can be seen that the motor still maintains independent operation at this time. Fig. 20 and by Fig.21 It can be seen that this method effectively reconstructs the three-phase current and can realize fault-tolerant control of the current sensor.

[0197] In order to verify the dynamic performance of the motor when the speed changes, a dynamic experiment of continuous step change of the speed of the upper motor is carried out. The load of the upper motor remains unchanged and the initial speed is set to 300r / min, which changes to 500r / min after 3s and 700r / min after 5s. The speed of the lower motor is 400r / min. The experimental waveform of the upper motor is as follows Figure 22-23 shown. Fig. 22 is the upper motor speed waveform, Fig.23 is the motor a phase current waveform.

[0198] According to the experimental results, the motor speed can be effectively tracked in real time, and the overshoot is less than 10%. When the speed is switched, the current has a small waveform.

[0199] In order to verify the dynamic performance of the motor when the load changes, a dynamic experiment of continuous step change of load is carried out on the upper motor. The upper motor constant speed is set to 300r / min, the initial load is 4N·m, which changes to 6N·m after 3s and 8N·m after 5s; the lower motor speed is 400r / min. The upper motor experimental waveform is as follows Figure 24-25 shown. Fig.24 is the upper motor speed waveform, Fig.25 It is the current waveform of phase a of the upper motor.

[0200] According to the experimental results, when the load changes, the speed is disturbed to a certain extent, but it can return to stability in time.

[0201] The experimental results show that the single current sensor phase current reconstruction method based on the nine-switch converter effectively reconstructs the phase current. The fault diagnosis algorithm can timely and effectively identify the normal and faulty states of the current sensor. The fault-tolerant control algorithm based on the finite set model predictive control can also effectively reconstruct the three-phase current. The constructed system basically meets the requirements, and the dynamic and static control performance is good.

[0202] The dual-motor drive system has the problems of a large number of power devices and current sensors, high cost, large size and weight, and low power density. The traditional single current sensor current reconstruction method based on the traditional three-phase half-bridge circuit has certain limitations.

[0203] In the present invention, the phase current reconstruction method of the dual permanent magnet synchronous motor single current sensor based on the nine-switch converter greatly reduces the number of power devices and current sensors. The current sensor fault diagnosis and fault-tolerant control based on the nine-switch model predictive control improves the reliability of the dual motor drive system based on the nine-switch converter. Moreover, the three-phase current fault-tolerant control based on the model predictive control is equivalent to the current sensor-free control, which further reduces the system cost, improves the power density, and improves the overall performance of the dual motor control.

[0204] The specific instructions are as follows:

[0205] 1. A single current sensor current sampling method is used in the nine-switch converter dual-motor drive system. Different strategies are used to reconstruct the current in the sector switching unobservable area, low modulation unobservable area, and high modulation unobservable area using phase shift method, voltage vector insertion method, and approximation method.

[0206] 2. Improved model predictive control based on nine-switch converter Fault-tolerant control To address the problem of three-phase current reconstruction, a new voltage vector table was established, and a current observer was constructed based on the switching function and the motor mathematical model. The finite set model predictive control was used to achieve independent control of dual motors, and the voltage vector screening method was used to improve it, which increased the operation speed within a single control cycle; the space vector distance function was introduced to improve the objective function and reduce the switching loss of the nine-switch converter;

[0207] 3. The present invention uses a method based on stator coordinate transformation to perform fault diagnosis on the current sensor of the nine-switch converter, and can diagnose the status of the current sensor;

[0208] 4. The present invention adopts a current reconstruction method of a dual permanent magnet synchronous motor with a single current sensor based on a nine-switch converter to realize closed-loop control of the dual motors. In order to improve reliability, a current sensor fault diagnosis and fault-tolerant control method for a dual permanent magnet synchronous motor drive system with a nine-switch converter is proposed.

[0209] So far, the technical solutions of the present invention have been described in conjunction with the preferred embodiments shown in the accompanying drawings. However, it is easy for those skilled in the art to understand that the protection scope of the present invention is obviously not limited to these specific embodiments. Without departing from the principle of the present invention, those skilled in the art can make equivalent changes or substitutions to the relevant technical features, and the technical solutions after these changes or substitutions will fall within the protection scope of the present invention.

Claims

1. A fault-tolerant reconstruction method for a dual permanent magnet synchronous motor based on a nine-switch converter, characterized in that: include: The three-phase current of the dual permanent magnet synchronous motor is reconstructed based on a nine-switch converter; wherein the nine-switch converter includes a first switch element, a second switch element, a third switch element, a fourth switch element, a fifth switch element, a sixth switch element, a seventh switch element, an eighth switch element, a ninth switch element, a resistor, a capacitor and a current sensor; Sampling the current of the synchronous motor based on the current sensor, and determining whether the current sensor fails; If a fault occurs, a current observer is constructed; Obtaining three-phase current values ​​of two motors based on the current observer; According to the obtained three-phase current values, fault-tolerant control is performed based on the improved model predictive control; The improved model predictive control includes: Obtaining voltage vector reference angles of the first motor and the second motor; Establishing a selection strategy according to the voltage vector reference angle, selecting 5 voltage vectors for traversal from 15 voltage vectors; Traversing five voltage vectors based on the objective function, selecting a voltage with a minimum value of the objective function; The switch state corresponding to the vector is used as the switch state of the next cycle; Voltage vector reference angle θ v * : i v * =θ e +d * +π / 2; In the formula, δ * is the power angle, δ * =arctan(ψ * q / ψ * d );According to the speed loop feedback i * q We can get ψ * d and ψ * q θ e is the stator electrical angle; The objective function: g=g1+λg2+μs; Among them, the objective function of the current prediction of the two motors is obtained by subtracting the dq-axis current at time k+1 from the q-axis current reference value and the d-axis reference current obtained by the two speed loops: In order to characterize the switching loss of the nine-switch converter, a space vector distance function is introduced and V is defined as k and V k+1 The space vector distance function between is as follows: λ is the weight coefficient of the two motors, and μ is the weight coefficient of the loss function.

2. The method according to claim 1, characterized in that The nine-switch converter comprises: The input end of the first switch element is connected to the first end of the capacitor, the output end of the first switch element is connected to the input end of the second switch element, the output end of the second switch element is connected to the input end of the third switch element, the control end of the first switch element receives a control signal to change the working state of the first switch element, the control end of the second switch element receives a control signal to change the working state of the second switch element, and the control end of the third switch element receives a control signal to change the working state of the third switch element; The input end of the fourth switch element is connected to the first end of the capacitor, the output end of the fourth switch element is connected to the input end of the fifth switch element, the output end of the fifth switch element is connected to the input end of the sixth switch element, the control end of the fourth switch element receives a control signal to change the working state of the fourth switch element, the control end of the fifth switch element receives a control signal to change the working state of the fifth switch element, and the control end of the sixth switch element receives a control signal to change the working state of the sixth switch element; The input end of the seventh switch element is connected to the first end of the capacitor, the output end of the seventh switch element is connected to the input end of the eighth switch element, the output end of the ninth switch element is connected to the input end of the sixth switch element, the control end of the seventh switch element receives a control signal to change the working state of the seventh switch element, the control end of the eighth switch element receives a control signal to change the working state of the eighth switch element, and the control end of the ninth switch element receives a control signal to change the working state of the ninth switch element; The output end of the third switch element, the output end of the sixth switch element and the output end of the ninth switch element are all connected to the first end of the resistor, and the second end of the resistor is connected to the second end of the capacitor; The first end and the second end of the resistor are respectively connected to the current sensor; The first end of the capacitor is used to be connected to the positive electrode of the power supply, and the second end of the capacitor is used to be connected to the negative electrode of the power supply; The output end of the first switching element is used to connect to the first phase line of the first motor, the output end of the fourth switching element is used to connect to the second phase line of the first motor, the output end of the seventh switching element is used to connect to the third phase line of the first motor, the output end of the second switching element is used to connect to the first phase line of the second motor, the output end of the fifth switching element is used to connect to the second phase line of the second motor, and the output end of the eighth switching element is used to connect to the third phase line of the second motor.

3. The method according to claim 1 or 2, characterized in that: The reconstructing the three-phase current of the dual permanent magnet synchronous motor based on the nine-switch converter includes: In the observable area and the non-observable area, the current sensor based on the nine-switch converter collects the three-phase current of the dual permanent magnet synchronous motor to complete the reconstruction of all phase currents. The collection method is to sample four times per cycle, and each collection obtains the two-phase currents of the two motors, for a total of four currents.

4. The method according to claim 3, characterized in that In the non-observation area, the three-phase current of the dual permanent magnet synchronous motor is collected by the current sensor of the nine-switch converter, including: When the voltage vector is located in the non-observation area of ​​sector switching, the action time of a single voltage vector is adjusted; When the voltage vector is located in the low modulation non-observation area, the action time of the two voltage vectors is adjusted. Only one voltage vector action time is adjusted in each cycle, and the action time of the other voltage vector adopts the action time of the previous cycle. When the voltage vector is located in the high modulation non-observation region, the voltage vector is adjusted based on an approximate processing method.

5. The method according to claim 4, characterized in that When the voltage vector is located in the sector switching non-observation area, adjusting the action time of a single voltage vector includes: The voltage vector action time of the first motor is T1, and the voltage vector action time of the second motor is T2. min is the minimum sampling time, T min =T d +T set +T adc ; T d Indicates dead time; T set Indicates the bus current establishment time; T adc Indicates the time required for sampling; When T1≥2T min And T min <T2<2T min When the voltage is equal to the input voltage, the action time of a single voltage vector is adjusted based on the phase shift method; When T1≥2T min And T2<2T min When the voltage vector is inserted, the action time of a single voltage vector is adjusted.

6. The method according to claim 1 or 2, characterized in that: The sampling of the current of the synchronous motor based on the current sensor to determine whether the current sensor fails includes: The current sensors are used to sample the current of each phase of the two motors to obtain the six-phase currents of the two motors; Obtaining a current residual of the sampled current; Based on the difference between the current residual and a preset threshold, it is determined whether the current sensor fails.

7. The method according to claim 6, characterized in that The determining whether the current sensor fails based on the difference between the current residual and a preset threshold value includes: If the current residual of any phase current is higher than a preset threshold, it is determined that the current sensor is faulty.