Self-adaptive open-phase fault-tolerant control method suitable for double-star delta winding permanent magnet synchronous motor under different phase shift angles

Abstract

The invention discloses a self-adaptive open-phase fault-tolerant control method suitable for a double-star delta winding permanent magnet synchronous motor under different phase shift angles, and ensures that the motor of the type can be automatically transited to a normal operation state when a current sensor fault or a winding open-circuit fault occurs in any phase of star winding. The method comprises the following steps: establishing a VSD transformation matrix of the double-star delta winding permanent magnet synchronous motor under different phase shift angles and a motor mathematical model based on a shaft and a shaft plane; analyzing a corresponding relation of each phase current when the star winding has an open-circuit fault, and calculating a relational expression between axis current and axis current based on a fundamental wave magnetomotive force invariant principle; establishing a self-adaptive default phase fault tolerance process; determining a first-order hyper-local model form of a motor mathematical model based on the shaft and the shaft plane; designing an extended state observer to estimate current and total disturbance quantity; and discretizing the first-order hyper-local model combined with the extended state observer to obtain a model-free predictive current control method and realize default phase fault tolerance.

Description

Technical Field

[0001] This invention relates to the field of fault-tolerant control technology for motors, and in particular to an adaptive phase-loss fault-tolerant control method for permanent magnet synchronous motors with double-star delta windings under different phase shift angles. Background Technology

[0002] In recent years, with the rapid development of fields such as ship propulsion, new energy vehicles, and aerospace, higher requirements have been placed on the power density, reliability, and fault tolerance of motors. Multiphase motors, with their advantages of power phase separation, low-voltage high-power output, high reliability, and flexible fault-tolerant control, have shown broad application prospects in these fields. Among them, the double-star delta winding permanent magnet synchronous motor, as a special multiphase motor topology, has the same number of driver bridge arms as the common double three-phase star winding motor, resulting in essentially the same drive hardware cost. However, it performs better in terms of electromagnetic torque density, torque ripple, permanent magnet eddy current losses, and vibration, demonstrating significant comprehensive economic benefits and performance advantages. It is considered one of the technical solutions with significant application potential.

[0003] However, as the number of phases and the required number of driver bridge arms increase, the probability of single-phase open-circuit faults in the motor also rises. Without fault tolerance, this can severely impact the motor's performance. When this type of motor is subjected to open-circuit fault-tolerant control, several control challenges arise: 1. The delta winding is located inside the motor and is not directly connected to the driver, making it impossible to independently and directly control the current of each phase like a star winding. This physical limitation makes it difficult to quickly reconstruct the current path of the delta winding under fault conditions. 2. There are issues of ambiguity and delay in the fault diagnosis process. When the winding current sampling feedback is consistently zero, it could be due to an open-circuit fault in the winding itself or a fault in the current sensor connected to that branch. Traditional methods typically require accurate diagnosis of the fault root cause before switching to the appropriate fault-tolerant mode, which inevitably prolongs the motor's fault operation time. 3. There is a strong magnetic field coupling effect between the two sets of star-delta windings inside the motor. It is difficult to directly measure accurate parameters such as inductance and resistance through the lead wires of the star winding at the end. Inaccurate parameters will seriously affect the accuracy and stability of the fault-tolerant control algorithm based on the internal parameter model of the motor. Summary of the Invention

[0004] (a) Technical problems to be solved

[0005] Based on this, the present invention provides an adaptive phase-loss fault-tolerant control method for a dual-star delta winding permanent magnet synchronous motor under different phase shift angles. For dual-star delta winding permanent magnet synchronous motors with special internal structures that are difficult to observe directly, when the motor experiences parameter input errors, single-phase current sensor failures, or single-phase winding open-circuit faults, an adaptive fault-tolerant control method is provided that can respond quickly, perform fault tolerance before diagnosis, and reduce the impact of motor parameters, ensuring that the motor can adaptively transition from the fault state to the normal operating state within one electrical cycle.

[0006] (II) Technical Solution

[0007] To achieve the above objectives, this invention provides an adaptive phase-loss-tolerant fault control method for a dual-star delta-winding permanent magnet synchronous motor applicable to different phase shift angles, comprising:

[0008] Step S1: Establish the VSD transformation matrix of the double-star delta winding permanent magnet synchronous motor under different phase shift angles and based on... shaft and A mathematical model of the motor in the axial plane; analysis of the corresponding relationships of the phase currents when an open-circuit fault occurs in the star winding, and calculation based on the principle of constant fundamental magnetomotive force. shaft and The formula for shaft current provides a basis for modifying the current reference value in subsequent fault-tolerant control.

[0009] Step S2: Establish an adaptive phase loss fault-tolerant process; specifically including:

[0010] Step S21: Establish the fault-tolerant transformation matrix ;

[0011] Step S22: Combining the current distribution characteristics of current sensor faults and winding open circuit faults, determine the adaptive fault-tolerant process of the motor when a phase loss fault occurs.

[0012] Step S23: Determine the criterion form for phase loss fault detection;

[0013] Step S24: Form a complete adaptive phase loss fault-tolerant process;

[0014] Step S3: Determine based on shaft and The first-order hyperlocal model of the motor mathematical model in the axial plane is presented; an extended state observer is designed to predict the current and total disturbance; the first-order hyperlocal model combined with the extended state observer is discretized to obtain a model-free predictive current control method, achieving phase loss tolerance.

[0015] (III) Beneficial Effects

[0016] As can be seen from the above technical solution, the adaptive phase-loss fault-tolerant control method for a dual-star delta winding permanent magnet synchronous motor proposed in this invention, applicable to different phase shift angles, has the following advantages:

[0017] 1. A systematic fault-tolerant theoretical foundation for double-star delta winding structures was established, providing a direct design basis for fault-tolerant control. A fault-tolerant analysis process for open-circuit faults in double-star delta winding permanent magnet synchronous motors under different phase shift angles was provided. Taking a single-phase star winding open-circuit fault as an example, the relationship between the currents of each phase winding under fault conditions was analyzed. After equivalently converting the delta winding current to the star winding, based on the principle that the fundamental magnetomotive force remains unchanged before and after the fault, a fault-tolerant control method applicable to both minimum copper loss control and maximum torque control was derived. shaft current and The relationship between shaft currents.

[0018] 2. An adaptive fault-tolerant control method with fault tolerance first and diagnosis later is proposed, which realizes "instant response" in fault handling and greatly improves the reliability of continuous system operation. That is, by combining the fault-tolerant transformation matrix and the phase loss fault-tolerant control method, it is ensured that when a single-phase current sensor fault or a single-phase open circuit fault occurs, the motor can automatically transition from the fault state to the normal operating state, and the fault operation time is kept within one electrical cycle.

[0019] 3. To reduce the impact of errors in motor parameter input on motor operating performance, this invention employs an extended model-free predictive current control strategy. shaft and The shaft current loop module replaces traditional PI control or deadbeat predictive current control. It reduces the dependence on precise motor parameters, significantly enhances the robustness and dynamic performance of fault-tolerant control, and has excellent current tracking and harmonic suppression capabilities. Attached Figure Description

[0020] The features and advantages of the invention will be more clearly understood by referring to the accompanying drawings, which are schematic and should not be construed as limiting the invention in any way. In the drawings:

[0021] Figure 1 This is a schematic diagram of the double-star delta winding permanent magnet synchronous motor and its drive topology according to the present invention;

[0022] Figure 2 This is a schematic diagram of a winding connection configuration commonly used in the prior art of this invention;

[0023] Figure 3 This is the first set of star-delta windings of the present invention. A schematic diagram of the current relationship during an open-circuit fault in a phase;

[0024] Figure 4This invention relates to a permanent magnet synchronous motor with a double-star delta winding under three different phase shift angles. Schematic diagram of current distribution under minimum copper loss control during phase open circuit fault;

[0025] Figure 5 This invention relates to a permanent magnet synchronous motor with a double-star delta winding under three different phase shift angles. Schematic diagram of current distribution under maximum torque control during a phase open-circuit fault;

[0026] Figure 6 This invention relates to a dual-star delta winding permanent magnet synchronous motor with three different phase shift angles under healthy and... Comparison of finite element torque output analysis of different measures taken during phase open circuit faults;

[0027] Figure 7 This is a schematic diagram of the equivalent replacement of the fault phase vector in this invention;

[0028] Figure 8 This is a flowchart of the adaptive fault-tolerant control of the present invention;

[0029] Figure 9 This is a block diagram illustrating the adaptive fault-tolerant control principle of the present invention for a permanent magnet synchronous motor with a double-star delta winding under different phase shift angles;

[0030] Figure 10 This is a physical image of the experimental platform for the dual-star delta winding permanent magnet synchronous motor drive system of the present invention;

[0031] Figure 11 This invention relates to a dual-star delta winding permanent magnet synchronous motor. When the phase current sensor fails, the proposed adaptive fault-tolerant method is applied to the speed, torque, and phase currents. Waveform diagram of the square root current and the flag bit;

[0032] Figure 12 A permanent magnet synchronous motor with double star delta winding During an open-circuit fault, minimum copper loss control and the proposed adaptive fault-tolerant method are used to control speed, torque, and phase current. Waveform diagram of the square root current and the flag bit;

[0033] Figure 13 This invention relates to a dual-star delta winding permanent magnet synchronous motor. During an open-circuit fault, maximum torque control and the proposed adaptive fault-tolerant method are used to control speed, torque, and phase current. Waveform diagram of the square root current and the flag bit;

[0034] Figure 14 This invention relates to a dual-star delta winding permanent magnet synchronous motor. When the phase current sensor fails, the proposed adaptive fault-tolerant method is applied to the speed, torque, and phase currents. Waveform diagram of the square root current and the flag bit;

[0035] Figure 15 This invention relates to a dual-star delta winding permanent magnet synchronous motor. During an open-circuit fault, minimum copper loss control and the proposed adaptive fault-tolerant method are used to control speed, torque, and phase current. Waveform diagram of the square root current and the flag bit;

[0036] Figure 16 This invention relates to a dual-star delta winding permanent magnet synchronous motor. During an open-circuit fault, maximum torque control and the proposed adaptive fault-tolerant method are used to control speed, torque, and phase current. Waveform diagram of the square root current and the flag bit;

[0037] Figure 17 The motor parameter input for this invention is 1.5. and Two different predictive current control methods under sudden open-circuit faults Waveforms of shaft reference current and actual current; where Figure (a) shows the waveforms under deadbeat predictive current control. The waveforms of the shaft reference current and the actual current are shown in Figure (b), which illustrates the current control without model prediction. Waveforms of shaft reference current and actual current;

[0038] Figure 18 The motor parameter input for this invention is 1.5. and Two different predictive current control methods are adopted under maximum torque control after an open-circuit fault. Waveforms of shaft reference current and actual current; where Figure (a) shows the waveforms under deadbeat predictive current control. The waveforms of the shaft reference current and the actual current are shown in Figure (b), which illustrates the current control without model prediction. Waveforms of shaft reference current and actual current;

[0039] Figure 19 For different motor parameters input in this invention and Two different predictive current control methods are adopted under minimum copper loss control after a phase open circuit fault. Comparison of THD content in phase current waveforms; where Figure (a) shows the THD content under deadbeat predictive current control. The phase current waveform THD content diagram, Figure (b) shows the THD content under model-free predictive current control. Phase current waveform and THD content diagram. Detailed Implementation

[0040] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0041] This invention provides an adaptive phase-loss-tolerant control method for a dual-star delta-winding permanent magnet synchronous motor under different phase shift angles, including:

[0042] Step S1: Establish the VSD transformation matrix of the double-star delta winding permanent magnet synchronous motor under different phase shift angles and based on... shaft and A mathematical model of the motor in the axial plane; analysis of the corresponding relationships of the phase currents when an open-circuit fault occurs in the star winding, and calculation based on the principle of constant fundamental magnetomotive force. shaft and The formula for shaft current provides a basis for modifying the current reference value in subsequent fault-tolerant control.

[0043] Double-star delta winding permanent magnet synchronous motor and its drive topology as follows Figure 1 As shown. The output lines of the two drivers are respectively connected to the external terminals of the star winding of the permanent magnet synchronous motor (…). Mutual Attainment (phase), while delta winding ( Mutual Attainment The phases are connected inside the motor. The phase shift angle between the two sets of star-delta windings is... .

[0044] Common winding configurations include: Figure 2 As shown, it includes star windings, delta windings, and star-delta windings. This invention relates to a permanent magnet synchronous motor with double star-delta windings (two sets of star-delta windings).

[0045] The mathematical model of a double-star delta-winding permanent magnet synchronous motor with different phase shift angles is decomposed into three orthogonal decoupling subspaces: , and For a plane, the corresponding VSD transformation matrix can be expressed as:

[0046]

[0047] Used from coordinate transformation to The Park transformation matrix of the coordinates can be represented as:

[0048]

[0049] in: This indicates the rotor electrical angle.

[0050] Because the star-delta winding has no neutral point, under normal operating conditions... The sum of plane currents is always zero, therefore it can be ignored. Mathematical model of an electric motor in a plane. (The formula is...) - Combining orthogonal decoupling analysis, the results of the double-star delta winding permanent magnet synchronous motor under different phase shift angles were obtained. shaft and The equations for voltage, flux linkage, torque, and speed in the axial plane (mathematical model of the motor) are expressed as follows:

[0051]

[0052]

[0053]

[0054]

[0055] in, , They represent In the axial plane axis, shaft voltage, , They represent In the axial plane axis, Shaft voltage; , They represent In the axial plane axis, shaft current, , They represent In the axial plane axis, shaft current; , They represent In the axial plane axis, Axis flux linkage variable, , They represent In the axial plane axis, Axis flux linkage variable; , They represent In the axial plane axis, Shaft inductance; Indicates leakage; , These represent the electrical angular frequency and mechanical angular frequency of the rotor, respectively. , These represent electromagnetic torque and load torque, respectively. Indicates permanent magnet flux linkage; Indicates the stator winding resistance; Represents the extreme logarithm; Indicates the moment of inertia; This represents the damping coefficient.

[0056] Before a single-phase open-circuit fault occurs in a double-star delta-winding permanent magnet synchronous motor at different phase shift angles, the relationship between the delta winding current and the corresponding star winding current can be expressed as:

[0057]

[0058] in, and These represent the currents in the delta-connected and star-connected windings, respectively; subscripts. and These represent delta winding and star winding, respectively; subscript The range is 1 to 6, representing respectively Mutual Attainment Mutually.

[0059] Because the current amplitude and phase flowing through the delta winding and the star winding are different, the fundamental magnetomotive force generated by the delta winding must be calculated separately. The induced magnetomotive force generated by a double-star delta winding before a single-phase open-circuit fault at different phase shift angles can be expressed as:

[0060]

[0061] in, Indicates the number of turns in the star winding; This indicates the magnitude of the current in each phase of the star winding before the fault. and These represent the first and second windings in the star winding and delta winding, respectively. phase and Electrical angle between phases (e.g.: Indicating the delta winding In the phase and star winding (electric angle between phases) express Phase current phase angle; This indicates the winding space angle.

[0062] With the first set of star-delta windings Taking a single-phase open-circuit fault as an example, since there is no neutral point in the star-delta winding configuration, the current relationship of the external star winding can be expressed as follows:

[0063]

[0064] Because the impedance of each phase in a delta winding is the same, the voltage difference across the delta winding is equal. Because in a delta winding... Harmony The phase impedance is Twice the amount of the phase, flowing in The current in the phase is Harmony Twice the phase current, such as Figure 3 As shown. Therefore, regardless of whether a single-phase open-circuit fault occurs in the external winding, the relationship between the currents in the delta winding and the star winding can be expressed as:

[0065]

[0066] Therefore, the fundamental induced magnetomotive force generated by the double-star delta winding at different phase shift angles can be represented solely by the external star winding current:

[0067]

[0068]

[0069] To ensure To maintain the fundamental magnetomotive force unchanged before and after a single-phase open-circuit fault, it is necessary to optimize the current of the healthy phase (the phase without fault) to improve the motor's operating performance under fault conditions. Commonly used open-circuit fault-tolerant strategies include maximum torque control and minimum copper loss control, with optimization objectives as shown in formulas (13) and (14), respectively:

[0070]

[0071]

[0072] in, This represents the objective function for maximum torque control. This represents the objective function for minimizing copper loss control.

[0073] To obtain the phase current of the healthy phase under the two open-circuit fault-tolerant strategies, a Lagrange equation with equality constraints is established, as follows:

[0074]

[0075] in, Represent the objective function; Represents constrained Lagrange multipliers; Represent the constraint equations; This indicates the number of constraints (i.e., constraint equations).

[0076] With formula For the objective function, the formula is... , and As constraints, combined with the formula get When an open-circuit fault occurs in a phase, the current of the healthy phase under minimum copper loss control is:

[0077]

[0078] With formula For the objective function, the formula is... , and As constraints, combined with the formula get When an open-circuit fault occurs in a phase, the current of the healthy phase under maximum torque control is:

[0079]

[0080] Taking maximum torque control as an example, the formula is... and Multiplying them together, we can get... shaft and The axis current relationship is as follows:

[0081]

[0082] when When an open circuit fault occurs in a phase, The shaft current is no longer related to Axis current decoupling. This is achieved by maintaining the formula. The provisions of China shaft and The relationship between shaft currents can ensure that... Under open-circuit fault conditions, the fundamental magnetomotive force remains unchanged, and maximum torque control is performed. The analysis method for minimum copper loss control follows the same principle.

[0083] Therefore, when a single-phase open-circuit fault occurs in a dual-star delta-winding permanent magnet synchronous motor under different phase shift angles, the maximum torque control is obtained. shaft and The current relationship of the shaft is as follows:

[0084] exist phase or When an open circuit fault occurs in a phase, shaft and The current relationship of the shaft is as shown in the formula. As shown;

[0085] exist phase or When an open circuit fault occurs in a phase, shaft and The current relationship of the shaft is shown below:

[0086]

[0087] exist phase or When an open circuit fault occurs in a phase, shaft and The current relationship of the shaft is shown below:

[0088]

[0089] When a single-phase open-circuit fault occurs in a dual-star delta-winding permanent magnet synchronous motor under different phase shift angles, the minimum copper loss control is obtained. shaft and The current relationship of the shaft is as follows:

[0090] exist When an open circuit fault occurs in a phase, shaft and The current relationship of the shaft is shown below:

[0091]

[0092] exist When an open circuit fault occurs in a phase, shaft and The current relationship of the shaft is shown below:

[0093]

[0094] exist When an open circuit fault occurs in a phase, shaft and The current relationship of the shaft is shown below:

[0095]

[0096] exist When an open circuit fault occurs in a phase, shaft and The current relationship of the shaft is shown below:

[0097]

[0098] exist When an open circuit fault occurs in a phase, shaft and The current relationship of the shaft is shown below:

[0099]

[0100] exist When an open circuit fault occurs in a phase, shaft and The current relationship of the shaft is shown below:

[0101]

[0102] To analyze the feasibility of the proposed phase loss tolerance method when a single-phase open-circuit fault occurs in a dual-star-delta winding permanent magnet synchronous motor under different phase shift angles, a finite element simulation model was built for verification. The phase shift angles of the two sets of star-delta windings in the three 48-slot, 22-pole motor models were 7.5°, 15°, and 60°, respectively. The operating conditions of the motor under healthy, open-circuit fault-tolerant, and open-circuit fault-tolerant states were simulated by controlling the current input to the external circuit of the motor. Taking an open-circuit fault in a phase winding as an example, if the fault is not easily corrected, then in each phase current, The phase current is always 0, and the current distribution of the other phases is the same as that under healthy conditions; if the formula is used after a fault... The corresponding minimum copper loss control results in the current distribution of the double-star delta winding permanent magnet synchronous motor at different phase shift angles as follows: Figure 4 As shown ( Figure 4 (a), 4(b), and 4(c) represent phase shift angles of 7.5°, 15°, and 60°, respectively); if the formula is adopted after the fault... The corresponding maximum torque control results in the current distribution of the dual-star delta winding permanent magnet synchronous motor at different phase shift angles as follows: Figure 5 As shown ( Figure 5(a), 5(b), and 5(c) represent phase shift angles of 7.5°, 15°, and 60°, respectively.

[0103] The feasibility of the proposed open-circuit fault-tolerant strategy is verified by analyzing the motor torque output performance under different modes using finite element analysis. Figure 6 ( Figure 6 (a), 6(b), and 6(c) represent the motor torque output waveforms under different phase shift angles of 7.5°, 15°, and 60°, respectively. It can be seen that under different phase shift angles, the torque output waveforms of the double-star delta winding permanent magnet synchronous motor... When an open-circuit fault occurs in a phase winding, the proposed minimum copper loss control and maximum torque control methods are adopted. The output torque waveform highly overlaps with the torque waveform under healthy conditions, and the average torque is almost the same. However, the average torque output of the motor in the fault-tolerant state decreases significantly. Although the torque ripple of the two fault-tolerant strategies (minimum copper loss control and maximum torque control) is slightly increased compared to the healthy state, this is due to the additional magnetomotive force harmonics generated by the winding. However, the torque ripple suppression effect is significantly better than that in the fault-tolerant state.

[0104] Step S2: Establish an adaptive phase loss fault-tolerant process; specifically including:

[0105] Step S21: Establish the fault-tolerant transformation matrix ;

[0106] When a single current sensor fails, the actual current in the faulty phase winding is not zero, but the sampled feedback current is zero. Since the star-delta winding has no neutral point, the actual sum of the currents in the three-phase star winding is zero. Therefore, the faulty phase component can be replaced by the healthy phase component to avoid the influence of the current sensor failure. Furthermore, considering that in the maximum torque control fault-tolerant mode, one phase current in each of the two star-delta windings will approach zero, leading to a false fault scenario, the faulty and false faulty phases in both pairs of star-delta windings need to be replaced by normal healthy phases. For example, suppose in… In the event of an open-circuit fault in the phase, under the maximum torque control fault-tolerant mode, Harmony The phase current is close to zero. The phase is the faulty phase. The phase is a pseudo-fault phase.

[0107] Based on this, or Taking a phase fault as an example, an equivalent vector distribution is established through vector substitution. The substitution process is as follows: Figure 7 As shown. By performing VSD transformation based on the new phase current vector distribution, accurate values ​​can be obtained. shaft and Axis current. Fault-tolerant transformation matrix. Represented as:

[0108]

[0109] Using formula The original 6D VSD transformation matrix containing the fault phase Convert to a normal 4D fault-tolerant transformation matrix The resulting transformation shaft and The shaft current no longer contains a fault component, thus avoiding... or The impact of a faulty current sensor in the phase on motor operation is investigated, while maintaining the effectiveness of the proposed open-phase fault-tolerant control.

[0110] The same principle applies to the fault-tolerant transformation matrix for other phase faults: for or In the event of a phase fault, the VSD transformation matrix can be switched to a fault-tolerant transformation matrix. , and for or In the event of a phase fault, the VSD transformation matrix can be switched to a fault-tolerant transformation matrix. .

[0111] Step S22: Combining the current distribution characteristics of current sensor faults and winding open circuit faults, determine the adaptive fault-tolerant process of the motor when a phase loss fault occurs.

[0112] When a motor experiences a phase loss fault, it needs to switch to the appropriate fault-tolerant transformation matrix in a timely manner. First, the number of faulty phases must be determined, which can be done by analyzing the root mean square value of the current sampling feedback. If the fault is caused by a current sensor malfunction, then switch to the appropriate fault-tolerant transformation matrix. The shaft current will rapidly decay to near zero. If the fault is caused by an open-circuit fault in the winding, even after switching to the corresponding fault-tolerant transformation matrix, The amplitude of the shaft current is still very large. At this point, adjustments can be made according to the open-circuit fault-tolerant control strategy. A shaft current reference is used for fault tolerance. It is used to determine the current feedback status of each phase and after switching the fault-tolerant transformation matrix. The expression for the change in shaft current is:

[0113]

[0114] in, This represents the RMS value of the current feedback for each phase; express The RMS value of the square root current; Indicates the time allocation coefficient; This represents one electrical cycle.

[0115] This invention primarily addresses the issue of a single-phase feedback current that is constantly zero, a phase loss fault that may be caused by a current sensor malfunction or an open-circuit fault. However, when a single switching device in the driver fails, the current may approach zero within half an electrical cycle. To reduce the fault phase detection time and avoid misdiagnosis caused by a single switching device failure in the driver, The value must be greater than 0.5, and the overall fault diagnosis time must be limited to one electrical cycle.

[0116] Step S23: Determine the criterion form for phase loss fault detection;

[0117] Under ideal operating conditions, a normally functioning motor exhibits near-zero emissions. Shaft current. When an open-circuit fault occurs in the winding... shaft current and Axial current coupling, and synthesis The magnitude of the shaft current vector is no longer zero. During fault detection, the corresponding fault-tolerant transformation matrix is ​​switched and changed. The criterion for the shaft current reference value is:

[0118]

[0119]

[0120] in, and Both represent threshold values.

[0121] Regarding thresholds Its main function is to determine whether the current is "actually zero". Because factors such as disturbances and D / A sampling errors can cause the detected current value in the faulty phase to be less than ideally zero, therefore... It should be set to a value slightly higher than the RMS value of the current residual caused by the aforementioned disturbance, typically 1.5 to 2 times that RMS value, to ensure that it does not trigger falsely during normal, minor fluctuations, while still being sensitive enough to detect actual faults. (When the formula...) In When the criterion is true, the VSD transformation matrix is ​​replaced with the corresponding fault-tolerant transformation matrix. For example, when When switching from VSD transformation matrix to fault-tolerant transformation matrix. .

[0122] Regarding thresholds Its setting is based on the theoretical amplitude of specific current components when the motor experiences different types of faults. When the motor experiences a single-phase winding open circuit fault, and When the shaft voltage output has a finite amplitude constraint, according to the formula to Any phase The RMS value of the square root current is 0.7071. However, if only a single current sensor fails while the power circuit is functioning normally, the corresponding RMS value for that phase will approach zero. Therefore, The theoretical range is 0 to 0.7071. Between these values, 0.3536 can be selected. As The value of .

[0123] Step S24: Form a complete adaptive phase loss fault-tolerant process;

[0124] Adaptive phase loss fault tolerance process as follows Figure 8 As shown. When the system detects the first... The phase current feedback is constant at 0 and the duration exceeds When, the phase is triggered Criterion (formula) ), and switch the corresponding fault-tolerant transformation matrix. .exist After the delay, trigger Criterion (formula) If the current sensor is faulty, then... The judgment is negative because... The shaft current can be limited to zero. However, in the case of an open-circuit fault, this limitation cannot be enforced. The shaft current is zero, which leads to If the criterion result is true, the system will then modify the settings according to the open-path fault-tolerance strategy. A shaft reference value is used for phase loss tolerance to ensure normal motor operation. Specifically, this includes:

[0125] Step (1): System read to The value; when the first Phase current feedback value ( to The current sample value is constant at 0 and the duration exceeds When, the phase is triggered Criterion;

[0126] Step (2): If the first phase If the criterion is true, then the VSD transformation matrix is ​​switched to the corresponding fault-tolerant transformation matrix. ;exist After the delay, trigger Criterion;

[0127] like or If true, then switch the VSD transformation matrix to a fault-tolerant transformation matrix. ;like or If true, then switch the VSD transformation matrix to a fault-tolerant transformation matrix. ;like or If true, then switch the VSD transformation matrix to a fault-tolerant transformation matrix. ;

[0128] Step (3): If If the criterion is true, then modify the control scheme according to the open-circuit fault-tolerant strategy (maximum torque control or minimum copper loss control). The current reference value of the shaft is used for phase loss fault tolerance to ensure the normal operation of the motor.

[0129] Step S3: Determine based on shaft and The mathematical model of the motor in the axial plane is presented in the form of a first-order hyperlocal model. An extended state observer is designed to predict the current and total disturbance. The first-order hyperlocal model combined with the extended state observer is discretized to obtain a model-free predictive current control method, achieving phase loss tolerance. Specifically, this includes:

[0130] Step S31: Construct a first-order hyperlocal model;

[0131] For a single-input, single-output system, all non-input and non-output quantities are treated as lumped disturbance terms, and the first-order hyperlocal model can be expressed as:

[0132]

[0133] in, and These represent the system's output and input, respectively. This represents the lumped disturbance term, which includes both known and unknown disturbances. This represents the adjustment parameters associated with the system.

[0134] Combining formulas ,formula and It can be rewritten as:

[0135]

[0136] Among them, subscript Represents subspace components, with values ​​of , , or ;For example, express shaft current;

[0137] Total disturbance It can be represented as:

[0138]

[0139] in, , , and All of these indicate unknown disturbances.

[0140] Step S32: Design the extended state observer;

[0141] To obtain the formula Input current in Total disturbance Construct an extended state observer, represented as:

[0142]

[0143] in, and This represents the feedback gain of the extended state observer; and They represent and The estimated value.

[0144] By analyzing the formula Perform Laplace transform, and The transfer function can be expressed as:

[0145]

[0146] Formula The root constraints of the characteristic equation in At this point, the value of the extended state observer feedback gain coefficient can be expressed as:

[0147]

[0148] Step S33: Discretize the first-order hyperlocal model combined with the extended state observer to obtain the model-free predictive current control method;

[0149] To obtain the required voltage vector, the formula will be used. Discretization. Due to the controller's one-cycle delay, it must be done in the... Calculate the first control cycle. The voltage vector for each control cycle, used for one-cycle delay compensation, can be expressed as:

[0150]

[0151] in, Indicates the control cycle; Indicates the current reference value; and These represent the current and next control cycles, respectively.

[0152] Through discrete formula In the formula middle The observed current vector and total disturbance vector at each control cycle point can be expressed as:

[0153]

[0154] Using the current vector and total disturbance observed by the extended state observer, the formula is... Discretization yields the current vector and total disturbance required for the next control cycle, ensuring stable and rapid current tracking of the given reference value, thus enabling the output of the desired voltage vector. (See formula...) As shown, the observed current and total disturbance are independent of motor parameters such as resistance and flux linkage. Only the parameters... It includes inductance and leakage inductance, but the parameter input error will be adjusted as a disturbance, which has a certain degree of error disturbance robustness and is suitable for star-delta winding motor systems where it is difficult to obtain accurate motor parameters.

[0155] The principle block diagram of the adaptive phase loss fault-tolerant control method of the present invention is as follows: Figure 9 As shown. During normal operation of the motor, shaft and The shaft current reference value is set to 0. The shaft current reference value is provided by the speed loop PI regulator. Both the CPF and OPT flags are kept at 0. The current of each phase of the motor is obtained by sampling through a current sensor and calculated using the formula... and Perform VSD and Park transforms to obtain the current time step. , , and After the shaft current is measured, the current vector and total disturbance for the next control cycle are obtained through an extended state observer. A first-order hyperlocal model then follows the specified current reference value. The output voltage vector is applied to the motor through coordinate transformation and a three-phase decoupled SVPWM modulation strategy. If the first... If a phase loss fault occurs (current sensor failure or open circuit fault), the formula will be triggered. Criteria in Change the CPF flag from 0 to Simultaneously switch the corresponding fault-tolerant transformation matrix To avoid the impact of current sensor failure. In the case of an open-circuit fault, the formula... Criteria in It is also triggered, changing the OPT flag from 0 to Then modify the first-order hyperlocal model. A shaft current reference value is used to achieve open-circuit fault tolerance. Among them, The value ranges from 1 to 6, representing respectively Mutual Attainment Mutually.

[0156] To verify the effectiveness of the model-free predictive current control method for adaptive phase loss tolerance, an experimental platform was built based on a 24-slot / 10-pole double-star delta-winding permanent magnet synchronous motor, as follows: Figure 10 As shown. The experimental system mainly consists of a controller (with a TMS320C28346 as the main processor), a driver, a DC power supply, a PC, an oscilloscope, resistors, a torque sensor, a load motor, and a dual-star-delta winding permanent magnet synchronous motor. The motor under test contains two sets of star-delta windings with a phase shift angle of 15°, and the load motor consumes power through a resistor. In the experiment, the driver switching frequency was set to 10 kHz, the DC power supply output voltage was 200 V, and the resistor value was 20 Ω. Formula Parameters in Set the value to 0.75 and determine the threshold. , .also, The shaft output voltage is limited to 20 V, according to the formula. middle Set it to 4000.

[0157] When running at steady state at 500 rpm Harmony Taking a sudden phase failure as an example, the effectiveness of the proposed adaptive fault-tolerant strategy for current sensor failures and open-circuit failures was verified through experimental analysis. The analysis included the rotational speed before and after the failure (…). Torque ), actual phase current ( to ), Root-square current ( ), sampling current ( ) and fault tolerance flags (OPT, CPF).

[0158] when When a phase current sensor fails, the proposed adaptive fault-tolerant strategy is used for fault-tolerant control, and the results are as follows: Figure 11 As shown. Before the fault occurred, the motor speed was 499.4 rpm, the output torque was 1.21 N·m, and the phase current amplitude was 1.23 A. The square root current amplitude of the shaft is 0.07A. After the fault occurred, The phase sampling current drops to zero, while the actual current remains non-zero. This difference leads to the calculated... The shaft current undergoes transient changes. Although it can track the given reference value in a timely manner under model-free predictive control, in reality... The loss of the phase current signal means that the calculated The shaft current no longer matches the actual current applied to the motor, thus affecting the motor's torque, speed, and other performance characteristics. For example... Figure 11 (a) and Figure 11 As shown in (c), after the fault occurs (i.e., to the right of the dashed line indicating the fault occurs in the figure), the torque fluctuation reaches 1.3 N·m, and the maximum current fluctuation reaches 2.6 A. 17 ms after the fault occurs, when... When the effective value of the phase sampling current is less than 0.15, the diagnostic flag bit CPF is set to 1, and the current VSD transformation matrix is ​​switched to... Thus eliminating The impact of a phase current sensor malfunction. Subsequently, as... Figure 11 (a) and Figure 11 As shown in (b), the torque, speed and current of each phase have all returned to normal operating conditions.

[0159] when When an open-circuit fault occurs, the control result using the proposed adaptive fault-tolerant strategy is as follows: Figure 12 and Figure 13 As shown. After the fault occurred, due to shaft current and The shaft current is no longer decoupled. The shaft current reference value is no longer zero. Forcing it to zero will cause severe torque ripple. Figure 12 (a) and Figure 13 As shown in (a), the motor torque ripple is 0.59 N·m. Figure 12 (c) and Figure 13 In (c), after 18ms and 17ms respectively following the occurrence of the fault, the system detected a continuous fault. Phase fault (CPF=1). The difference in current amplitude at the moment of fault occurrence caused the observed change in detection time. After 6ms, based on The shaft current feedback executes the specified open-circuit fault-tolerant control (OPT=1). Subsequently, the motor torque and speed return to normal operating conditions. When minimum copper loss control is used, such as... Figure 12(a) and Figure 12 As shown in (b) Mutual Attainment The phase current amplitudes are 0A, 1.07A, 1.07A, 2.38A, 1.93A, and 1.34A, respectively, which correspond to 0, 0.87, 0.87, 1.93, 1.57, and 1.09 times the peak current before the fault, satisfying the formula... When using maximum torque control, such as Figure 13 (a) and Figure 13 As shown in (b) Mutual Attainment The phase current amplitudes are 0A, 2.22A, 2.24A, 2.19A, 2.07A and 0.18A, respectively. The non-zero values ​​of the phase currents originate from model-free predictive current control. The inherent tracking error when the shaft current tracks its AC reference value. These amplitudes correspond to 0, 1.80, 1.82, 1.78, 1.68, and 0.15 times the peak current before the fault, respectively, satisfying the formula... It can be observed that after the fault occurs, the copper loss under minimum copper loss control is 0.71 times that under maximum torque control, while its maximum current amplitude is 1.05 times that of the latter.

[0160] To verify the universality of the proposed adaptive fault-tolerant strategy, we now examine its application in motors. Effectiveness when the phase current sensor fails or an open circuit fault occurs. When the phase current sensor fails, the result is as follows: Figure 14 As shown. The waveform changes of torque, speed, current, etc. before and after the fault are compared with... Figure 11 Similar, only the fault flag bits are different. For example... Figure 14 As shown in (c), 17ms after the fault occurred, the CPF changed from 0 to 5, indicating that... The phase current feedback value is always zero. When When an open-circuit fault occurs in a phase, the result is as follows: Figure 15 and Figure 16 As shown. 6ms after the CPF flag is updated, the OPT flag changes from 0 to 5, which is achieved by adjusting... Open-circuit fault-tolerant control is achieved using the shaft current reference value. Under minimum copper loss control, such as... Figure 15 (a) and Figure 15 As shown in (b) The phase current amplitude is maximum, reaching 2.34A. Under maximum torque control, such as... Figure 16 (a) and Figure 16 As shown in (b) The phase current amplitude is the smallest, at 0.16A (close to zero), while the current amplitudes of the other healthy phases are similar, with an average value of about 2.17A.

[0161] Experimental results show that the proposed adaptive fault-tolerant method combining model-free predictive current control can autonomously switch fault-tolerant strategies to maintain normal motor operation when current sensor faults or open-circuit faults occur, demonstrating strong adaptive capability. Furthermore, the detection time for current sensor faults is shortened to [missing information]. This reduces the operating time of the motor under fault conditions.

[0162] Still with Taking an open-circuit fault as an example, combined with Figures 17 to 19 The experimental results shown, compared with the deadbeat predictive current control strategy, fully verify the effectiveness of model-free predictive current control in situations where motor inductance parameters are inaccurate (e.g., ...). , Superior performance under conditions where the deviation is 1.5 times the model value. Figure 17-18 This indicates that, under single-phase open-circuit fault conditions, the model-predictive current control strategy... shaft and Axis current tracking error (e.g.) The shaft current error increased only from 0.05A to 0.18A, which is significantly lower than that of deadbeat predictive current control. The shaft current error increased from 0.24A to 0.36A. (In the figure, the subscript...) Indicates the current reference value, for example: express Shaft current reference value. Also... Figure 19 This indicates that, under the same parameter deviation, the phase current harmonic distortion is smaller under the model-free predictive current control strategy, and the waveform quality is significantly better than that of the deadbeat predictive current control. Secondly, the extended model-free predictive current control strategy does not depend on the motor flux linkage and resistance parameters, making it particularly suitable for applications where the parameters of a dual-star winding permanent magnet synchronous motor are difficult to obtain precisely, significantly improving the system's fault tolerance robustness and operational reliability under parameter uncertainties.

[0163] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in the present invention should be included within the scope of protection of the present invention.

Claims

1. An adaptive phase-loss-tolerant fault control method for a dual-star delta-winding permanent magnet synchronous motor applicable to different phase shift angles, characterized in that, include: Step S1: Establish the VSD transformation matrix of the double-star delta winding permanent magnet synchronous motor under different phase shift angles and based on... shaft and A mathematical model of the motor in the axial plane; analysis of the corresponding relationships of the phase currents when an open-circuit fault occurs in the star winding, and calculation based on the principle of constant fundamental magnetomotive force. shaft and The formula for shaft current provides a basis for modifying the current reference value in subsequent fault-tolerant control. Step S2: Establish an adaptive phase loss fault-tolerant process; specifically including: Step S21: Establish the fault-tolerant transformation matrix ; Step S22: Combining the current distribution characteristics of current sensor faults and winding open circuit faults, determine the adaptive fault-tolerant process of the motor when a phase loss fault occurs. Step S23: Determine the criterion form for phase loss fault detection; Step S24: Form a complete adaptive phase loss fault-tolerant process; Step S3: Determine based on shaft and The first-order hyperlocal model of the motor mathematical model in the axial plane is presented; an extended state observer is designed to predict the current and total disturbance; the first-order hyperlocal model combined with the extended state observer is discretized to obtain a model-free predictive current control method, achieving phase loss tolerance.

2. The method according to claim 1, characterized in that, In step S1, the mathematical model of the double-star delta-winding permanent magnet synchronous motor with different phase shift angles is decomposed into three orthogonal decoupling subspaces: , and The plane, and the corresponding VSD transformation matrix, are represented as: Used from coordinate transformation to The Park transformation matrix of the coordinates is represented as: Where: θ e Indicates the rotor electrical angle; Because the star-delta winding has no neutral point, under normal operating conditions... The sum of plane currents is always zero, therefore it can be ignored. A mathematical model of the motor in the plane is derived; combined with orthogonal decoupling analysis, the results of the double-star delta winding permanent magnet synchronous motor under different phase shift angles are obtained. shaft and The equations for voltage, flux linkage, torque, and speed in the axial plane, i.e., the mathematical model of the motor, are expressed as follows: in, , They represent In the axial plane axis, shaft voltage, , They represent In the axial plane axis, Shaft voltage; , They represent In the axial plane axis, shaft current, , They represent In the axial plane axis, shaft current; , They represent In the axial plane axis, Axis flux linkage variable, , They represent In the axial plane axis, Axis flux linkage variable; , They represent In the axial plane axis, Shaft inductance; Indicates leakage; , These represent the electrical angular frequency and mechanical angular frequency of the rotor, respectively. , These represent electromagnetic torque and load torque, respectively. Indicates permanent magnet flux linkage; Indicates the stator winding resistance; Represents the extreme logarithm; Indicates the moment of inertia; Indicates the damping coefficient; Before a single-phase open-circuit fault occurs in a double-star delta-winding permanent magnet synchronous motor at different phase shift angles, the relationship between the delta winding current and the corresponding star winding current is expressed as follows: in, and These represent the currents in the delta-connected and star-connected windings, respectively; subscripts. and These represent delta winding and star winding, respectively; subscript The range is 1 to 6, representing respectively Mutual Attainment Mutually; Because the current amplitude and phase flowing through the delta winding and the star winding are different, the fundamental magnetomotive force generated by the delta winding must be calculated separately; the induced magnetomotive force generated by the double-star delta winding before a single-phase open-circuit fault at different phase shift angles is expressed as: in, Indicates the number of turns in the star winding; This indicates the magnitude of the current in each phase of the star winding before the fault. and These represent the first and second windings in the star winding and delta winding, respectively. phase and Electrical angle between phases; express Phase current phase angle; Indicates the winding space angle; When the first set of star-delta windings When a single-phase open-circuit fault occurs, since there is no neutral point in the star-delta winding configuration, the current relationship of the external star winding is expressed as follows: Because the impedance of each phase of the delta winding is the same, the voltage difference across the delta winding is equal; because the delta winding... Harmony The phase impedance is Twice the amount of the phase, flowing in The current in the phase is Harmony The current is twice that of the phase current; therefore, regardless of whether a single-phase open-circuit fault occurs in the external winding, the relationship between the currents in the delta winding and the star winding is expressed as: Therefore, the fundamental induced magnetomotive force generated by the double-star delta winding at different phase shift angles can be represented using only the external star winding current: To ensure To maintain the fundamental magnetomotive force unchanged before and after a single-phase open-circuit fault, it is necessary to optimize the current of the healthy phase to improve the motor's operating performance under fault conditions. Commonly used open-circuit fault-tolerant strategies include maximum torque control and minimum copper loss control, and their optimization objectives are as follows: in, This represents the objective function for maximum torque control. This represents the objective function for minimizing copper loss control. To obtain the phase current of the healthy phase under the two open-circuit fault-tolerant strategies, a Lagrange equation with equality constraints is established, as follows: in, Represent the objective function; Represents constrained Lagrange multipliers; Represent the constraint equations; Indicates the number of constraint equations; by Given the objective function, and combining the constraints and the Lagrange equation, we obtain... When an open-circuit fault occurs in a phase, the current of the healthy phase under minimum copper loss control is: by Given the objective function and constraints, and combining the constraints with the Lagrange equation, we obtain... When an open-circuit fault occurs in a phase, the current of the healthy phase under maximum torque control is: Under maximum torque control, the following is obtained: When an open circuit fault occurs shaft and The axis current relationship is as follows: when When an open circuit fault occurs in a phase, The shaft current is no longer related to Shaft current decoupling; by maintaining the conditions specified in the above formula. shaft and The relationship between the shaft currents ensures that... Under open-circuit fault conditions, the fundamental magnetomotive force remains unchanged and maximum torque control is performed; Therefore, when a single-phase open-circuit fault occurs in a dual-star delta-winding permanent magnet synchronous motor under different phase shift angles, the maximum torque control is obtained. shaft and The current relationship of the shaft is as follows: exist phase or When an open circuit fault occurs in a phase, shaft and The current relationship of the shaft is shown in the above formula; exist phase or When an open circuit fault occurs in a phase, shaft and The current relationship of the shaft is shown below: exist phase or When an open circuit fault occurs in a phase, shaft and The current relationship of the shaft is shown below: When a single-phase open-circuit fault occurs in a dual-star delta-winding permanent magnet synchronous motor under different phase shift angles, the minimum copper loss control is obtained. shaft and The current relationship of the shaft is as follows: exist When an open circuit fault occurs in a phase, shaft and The current relationship of the shaft is shown below: exist When an open circuit fault occurs in a phase, shaft and The current relationship of the shaft is shown below: exist When an open circuit fault occurs in a phase, shaft and The current relationship of the shaft is shown below: exist When an open circuit fault occurs in a phase, shaft and The current relationship of the shaft is shown below: exist When an open circuit fault occurs in a phase, shaft and The current relationship of the shaft is shown below: exist When an open circuit fault occurs in a phase, shaft and The current relationship of the shaft is shown below: 。 3. The method according to claim 2, characterized in that, In step S21, when a single current sensor fails, the actual current in the faulty phase winding is not zero, but the sampled feedback current is zero; since the star-delta winding has no neutral point, the actual sum of the currents in the three-phase star winding is zero. Therefore, the faulty phase component can be replaced by the healthy phase component to avoid the impact of current sensor failure; In addition, considering that in the maximum torque control fault-tolerant mode, the current in one phase of each of the two star-delta windings will be close to zero, resulting in a pseudo-fault scenario, it is necessary to replace the faulty phase and pseudo-faulty phase in the two pairs of star-delta windings with normal healthy phases. Based on this, when or In the event of a phase fault, an equivalent vector distribution is established through vector substitution; based on the new phase current vector distribution, a VSD transformation is performed to obtain an accurate... shaft and Axis current; fault-tolerant transformation matrix Represented as: The original 6D VSD transformation matrix containing the fault phase Convert to a normal 4D fault-tolerant transformation matrix The resulting transformation shaft and The shaft current no longer contains a fault component, thus avoiding... or The impact of a current sensor failure in one phase on motor operation is investigated, while maintaining the effectiveness of the proposed open-phase fault-tolerant control. The same principle applies to the fault-tolerant transformation matrix for other phase faults: for or In case of phase failure, switch the VSD transformation matrix to a fault-tolerant transformation matrix. , and for or In case of phase failure, switch the VSD transformation matrix to a fault-tolerant transformation matrix. ; In step S22, when a phase loss fault occurs in the motor, in order to switch to the corresponding fault-tolerant transformation matrix in a timely manner... First, the number of faulty phases must be determined, which is done by using the root mean square value fed back from the current sampling. If the fault is caused by a current sensor malfunction, then switch to the appropriate fault-tolerant transformation matrix. The shaft current will rapidly decay to near zero; If the fault is caused by an open-circuit fault in the winding, even after switching to the corresponding fault-tolerant transformation matrix... The amplitude of the shaft current is still very large; at this point, adjustments should be made according to the open-circuit fault-tolerant control strategy. The shaft current reference provides fault tolerance; it is used to determine the current feedback status of each phase and after switching the fault tolerance transformation matrix. The expression for the change in shaft current is: in, This represents the RMS value of the current feedback for each phase; express The RMS value of the square root current; Indicates the time allocation coefficient; This represents one electrical cycle; In step S23, under ideal operating conditions, a normally operating motor exhibits near-zero [efficiency / significance]. Shaft current; when an open-circuit fault occurs in the winding. shaft current and Axial current coupling, and synthesis The magnitude of the shaft current vector is no longer zero; during fault detection, the corresponding fault-tolerant transformation matrix is ​​switched and changed. The criterion for the shaft current reference value is: in, and Both represent threshold values; In step S24, when the system detects the first The phase current feedback is constant at 0 and the duration exceeds When, the phase is triggered Criterion, i.e. Check if it is true, and switch the corresponding fault-tolerant transformation matrix. ;exist After the delay, trigger Criterion, i.e. Is it true? If it's a current sensor malfunction, then... The judgment is negative because... The shaft current can be limited to zero; however, it cannot be forcibly constrained in the event of an open-circuit fault. The shaft current is zero, which leads to If the criterion result is true, the system will then modify the settings according to the open-path fault-tolerance strategy. The shaft reference value is used for phase loss tolerance to ensure normal motor operation; specifically including: Step (1): System read to The value; when the first Phase current feedback value, i.e. to The current sample value is constant at 0 and the duration exceeds When, the phase is triggered Criterion; Step (2): If the first phase If the criterion is true, then the VSD transformation matrix is ​​switched to the corresponding fault-tolerant transformation matrix. ;exist After the delay, trigger Criterion; like or If true, then switch the VSD transformation matrix to a fault-tolerant transformation matrix. ;like or If true, then switch the VSD transformation matrix to a fault-tolerant transformation matrix. ;like or If true, then switch the VSD transformation matrix to a fault-tolerant transformation matrix. ; Step (3): If If the criterion is true, then modify the open-path fault-tolerant strategy. The shaft current reference value is used for phase loss fault tolerance to ensure normal motor operation; among which, the open circuit fault tolerance strategy includes maximum torque control and minimum copper loss control.

4. The method according to claim 3, characterized in that, In step S23, due to factors such as disturbances and D / A sampling errors, the current value detected in the faulty phase is not ideally zero. Set to 1.5 to 2 times the RMS value of the current residual caused by the above disturbance.

5. The method according to claim 4, characterized in that, In step S23, The value range is from 0 to 0.7071. .

6. The method according to claim 5, characterized in that, In step S31, for a single-input single-output system, all non-input and non-output quantities are treated as lumped disturbance terms, and the first-order hyperlocal model is expressed as: in, and These represent the system's output and input, respectively. This represents the lumped disturbance term, which includes both known and unknown disturbances. Indicates the adjustment parameters associated with the system; Combining the above formula, the mathematical model of the motor can be rewritten as: Among them, subscript Represents subspace components, with values ​​of , , or ; Total disturbance Represented as: in, , , and Both indicate unknown disturbances; In step S32, in order to obtain the input current Total disturbance Construct an extended state observer, represented as: in, and This represents the feedback gain of the extended state observer; and They represent and The estimated value; By performing a Laplace transform on the above equation... and The transfer function is expressed as: Constrain the roots of the characteristic equation in the above equation. At this point, the value of the extended state observer feedback gain coefficient is expressed as: In step S33, in order to obtain the required voltage vector, the re-representation of the motor mathematical model is discretized; due to the controller's one-step delay, it must be done in the first step. Calculate the first control cycle. The voltage vector for each control cycle is used for one-cycle delay compensation, and is expressed as: in, Indicates the control cycle; Indicates the current reference value; and These represent the current and next control cycles, respectively. By discretizing the extended state observer, in the above equation The observed current vector and total disturbance vector at each control cycle point are expressed as follows: By discretizing the current vector and total disturbance observed by the extended state observer, the current vector and total disturbance required for the next control cycle are obtained, ensuring that the current stably and quickly tracks the given reference value, thus enabling the output of the desired voltage vector.

7. The method according to claim 6, characterized in that, In step S3, during normal operation of the motor, shaft and The shaft current reference value is set to 0. The shaft current reference value is provided by the speed loop PI controller. The CPF and OPT flags are both kept at 0. The current of each phase of the motor is obtained by sampling through current sensors, and the current at the current moment is obtained through VSD transformation and Park transformation. , , and After the shaft current is measured, the current vector and total disturbance of the next control cycle are obtained through the extended state observer. A first-order hyperlocal model is used to follow the specified current reference value, and the output voltage vector is applied to the motor through coordinate transformation and a three-phase decoupled SVPWM modulation strategy. If the first... If a phase loss fault occurs, the criterion will be triggered. Change the CPF flag from 0 to Simultaneously switch the corresponding fault-tolerant transformation matrix To avoid the impact of current sensor failure; in the case of an open circuit fault, the criterion is... It is also triggered, changing the OPT flag from 0 to Then modify the first-order hyperlocal model. Shaft current reference value to achieve open-circuit fault tolerance; among which, The value ranges from 1 to 6, representing respectively Mutual Attainment Mutually.

Images