Phase-loss self-fault-tolerant control method for phase-shift-free dual three-phase permanent magnet synchronous motor

Abstract

The invention discloses an open-phase self-fault-tolerant control method for a phase-shift-free dual three-phase permanent magnet synchronous motor. The method can automatically adjust a current distribution relation in the case of an open-phase fault and suppress torque pulsation in combination with neutral point voltage compensation. Comprising the following steps: driving two pairs of three-phase winding modules of the phase-shift-free DTP-PMSM in parallel by adopting two two-level three-phase voltage source inverters, and providing a mathematical model of the motor; current distribution is automatically adjusted; compensating the voltage of the neutral point, and calculating the offset of the fault module; the current is redistributed to suppress the torque ripple after the fault, and the secondary oblique wave component generated by the phase failure fault is eliminated; a current loop controller is improved through dead-beat predictive control; the speed ring controller is optimally designed into an improved sliding mode controller; an improved double-loop controller system is formed by combining improved sliding mode control and dead-beat prediction control, high-gain parameters are set to improve the dynamic response performance of the system, and a new self-fault-tolerant control system is formed.

Description

Technical Field

[0001] This invention belongs to the field of permanent magnet synchronous motor control technology, specifically relating to a phase-loss self-tolerant fault control method for a phase-shift-free dual three-phase permanent magnet synchronous motor. Background Technology

[0002] Phase-shiftless dual three-phase permanent magnet synchronous motors (DTP-PMSMs) have demonstrated significant application value in high-reliability fields such as electric vehicles and aerospace due to their high power density and fault tolerance. Traditional DTP-PMSM drive schemes employ a dual-inverter independent topology, where two independent three-phase inverters control two winding modules of the motor respectively. While this structure provides basic fault tolerance, its high hardware redundancy leads to increased system size, significantly higher manufacturing costs, and increased complexity in circuit connections and potential failure points.

[0003] In handling phase loss faults, existing technologies heavily rely on external fault diagnosis modules for fault detection and mode switching. This proactive intervention mechanism inevitably introduces response delays, leading to increased torque ripple and deterioration of current harmonics after phase loss, thus affecting system stability. Furthermore, traditional control architectures commonly employ dual-loop proportional-integral (PI) controllers, which have significant limitations in dynamic performance: 1) the speed loop is prone to speed overshoot and recovery delay when facing sudden load changes or parameter disturbances; 2) the current loop struggles to effectively suppress nonlinear current distortion under phase loss conditions, further amplifying electromagnetic torque fluctuations.

[0004] While recent research has attempted to simplify hardware structures through single-inverter topologies, this often comes at the cost of sacrificing fault tolerance. Improvements using complex observers or adaptive algorithms, while partially enhancing dynamic performance, introduce computational burdens. Therefore, current technologies have not yet effectively addressed the three core issues of excessive hardware redundancy, phase-loss response delay, and insufficient dynamic robustness. An innovative solution that combines structural simplicity, delay-free self-fault tolerance, and strong disturbance rejection is urgently needed. Summary of the Invention

[0005] To overcome the shortcomings of existing technologies, the present invention aims to provide a technical solution for a phase-loss self-tolerant control method for a single inverter parallel drive dual three-phase permanent magnet synchronous motor based on an improved dual-loop controller (ISMC+DPCC). By improving the current loop and speed loop controllers and combining them with a single inverter parallel drive architecture, the invention achieves high dynamic response, low torque ripple, and low harmonic distortion control under phase-loss fault conditions.

[0006] The technical effects required by this invention can be achieved through the following specific technical solutions:

[0007] The aforementioned self-fault-tolerant phase-loss control method for a phase-shiftless dual three-phase permanent magnet synchronous motor includes the following steps:

[0008] Step 1: Two two-level three-phase voltage source inverters are connected in parallel to drive two pairs of three-phase winding modules of a phase-shiftless DTP-PMSM and a mathematical model of a three-phase permanent magnet synchronous motor is proposed.

[0009] Step 2: Automatically adjust the current distribution to make the α-axis and β-axis currents of the two pairs of three-phase winding modules in a specific ratio; compensate the neutral point voltage and calculate the offset of the faulty module; redistribute the current to suppress the torque pulsation after the fault and eliminate the secondary ramp component caused by the phase loss fault, so as to achieve the effect of self-fault tolerance.

[0010] Step 3: Improve the current loop controller by adopting deadbeat predictive control;

[0011] Step 4: The speed loop controller is optimized and redesigned as an improved sliding diaphragm controller;

[0012] Step 5: Combine improved sliding mode control and deadbeat predictive control to form an improved dual-loop controller system. Set high-gain parameters to improve the dynamic response performance of the system and form a new self-fault-tolerant control system.

[0013] Furthermore, in step 1, the two pairs of three-phase winding modules share the same DC bus voltage, and the line current is the sum of the phase currents from the two pairs of three-phase winding modules; therefore, the mathematical model of the motor is as follows:

[0014]

[0015] Among them, i dk i qk It is the current on the dq axis, k = 1 or 2; I m It is the sum of the amplitudes of the phase currents in the two three-phase winding modules; the output torque of the three-phase permanent magnet synchronous motor is the sum of the electromagnetic torques of the two three-phase winding modules, i d =0, output torque T e It can be represented as:

[0016] T e =1.5Pψ f (i q1 +i q2 (3)

[0017] Where, ψ f The flux linkage generated by the permanent magnet magnetic field; P represents the number of pole pairs of the motor; i q1 i q2 These are the q-axis currents of three-phase winding module I and three-phase winding module II, respectively;

[0018] Furthermore, in step 2, the torque in each three-phase winding module and the total motor torque are:

[0019]

[0020] Among them, T e1 T is the torque of the three-phase winding module I. e2 T represents the torque of the three-phase winding module II. e The torque is the total torque of the entire motor; I β It is the current in an orthogonal coordinate system, and it is a process quantity of coordinate change.

[0021] Furthermore, in step 3, a deadbeat predictive current controller is used to replace the traditional PI controller, the continuous voltage equation is discretized, and the ideal control voltage is calculated based on the current reference value at the next moment, thereby achieving high dynamic response and reducing harmonic distortion.

[0022] Furthermore, the state equations of the surface-mounted, phase-shift-free dual three-phase permanent magnet synchronous motor in the dp rotating coordinate system are as follows:

[0023]

[0024] Among them, u d u q The voltage component along the dq axis; L d L q The stator is equivalent to the inductance on the dq axis; ψ f For permanent magnet flux linkage; ω e The electric angle is in radians.

[0025] Because the sampling time T s Since the equation is short enough, the first-order Euler forward discretization method is used to discretize the above equation, resulting in:

[0026]

[0027] Substituting equation (19) into equation (18) and rewriting it in the form of a matrix equation:

[0028]

[0029] Based on the fundamental principle of deadbeat current prediction, predicting the desired voltage requires using the current given value as the current value at the next moment, i.e.:

[0030]

[0031] Combining equations (20) and (21), the ideal deadbeat control voltage is obtained as follows:

[0032]

[0033] Among them, u d * (k) represents the ideal control voltage along the d-axis at time k, calculated using a predictive model; u q * (k) represents the ideal control voltage along the q-axis at time k, calculated using a predictive model; i d * (k+1) is the reference value of the d-axis current at time k+1, given by the outer loop controller (ISMC); i q * (k+1) is the reference value of the q-axis current at time k+1, given by the outer loop controller (ISMC); ω e (k) is the electric angular velocity of the motor at time k, which is the core parameter for calculating the back electromotive force and coupling terms.

[0034] Furthermore, in step 4, an improved sliding mode control is designed, a novel sliding mode surface containing the integral of the speed error and a nonlinear function is constructed, and the sigmoid function is used to optimize the reaching law to suppress chattering; and a disturbance observer is integrated to construct a second-order model to estimate the disturbance, and a high-gain feedback mechanism is added to compensate for load mutations, thereby improving the dynamic response capability of the system.

[0035] Furthermore, in step 4, the sliding surface function is defined as:

[0036]

[0037] The new exponential reaching law based on the sigmoid function is as follows:

[0038]

[0039] Therefore, the reference current for the q-axis after the improved sliding film control can be obtained as follows:

[0040]

[0041] Among them, T L ω represents the load torque, characterizing the external resistance at the motor shaft end; J represents the rotor, reflecting the inertia of the mechanical system; m ε is the mechanical angular velocity, defining the actual output speed; k1 and k2 are the sliding surface control gains, adjusting the linear and nonlinear error compensation strengths respectively; M is the torque constant, establishing the current-torque conversion relationship; B is the viscous friction coefficient; x1 is the state variable, representing the speed tracking error; σ(s) is the sigmoid-derived switching component (dimensionless), replacing the traditional sign function to suppress chattering σ(s) = sigmoid(s); ε2 and q2 are the approach law control parameters (dimensionless), controlling the disturbance rejection strength and convergence speed respectively.

[0042] Furthermore, the control law for the motor is:

[0043]

[0044] Let the state variable z1 = ω m z2 = -d, output y = ω m Since the disturbance d in the actual motor system changes slowly, its first derivative can be considered as 0. Therefore, the state-space equation of the system is established as follows:

[0045]

[0046] Using z1 and z2 as the observation objects, a high-gain feedback mechanism for the rotational speed estimation error e is established, and the disturbance observer is designed as follows:

[0047]

[0048] In the formula, z1 is the actual input mechanical angular velocity; The mechanical angular velocity estimated by the observer; For the estimated matching perturbation, α1 and α2 are positive real numbers; through this perturbation observer, it is possible to achieve

[0049] a = -F / J, b = 3p n ψ f / 2J;u represents the control law to be designed;

[0050] As shown in equation (40), the load disturbance observed by the disturbance observer is compensated into the speed loop controller, thus yielding:

[0051]

[0052] The novel speed loop controller designed as shown in equation (41) is called the Improved Sliding Film Controller (ISMC).

[0053] Compared with the prior art, the present invention has the following advantages:

[0054] This invention employs two two-level three-phase inverters connected in parallel to drive a phase-shift-free dual three-phase permanent magnet synchronous motor, sharing the DC bus voltage, and establishing a mathematical model of the motor under phase-loss fault conditions. A deadbeat predictive current controller (DPCC) replaces the current loop PI controller, achieving high dynamic response through discretized voltage equations. An improved sliding mode controller (ISMC) replaces the speed loop PI controller, using a sigmoid function to construct a second-order reaching law to suppress chattering. During a phase-loss fault, the current distribution relationship is automatically adjusted, combined with neutral point voltage compensation to suppress torque ripple. Compared to traditional phase-loss self-tolerant control strategies, this further reduces the negative impact of phase-loss faults on the system, improves the fault-tolerant performance of the motor system after a phase-loss fault, significantly enhances the anti-interference performance and dynamic response performance of the motor system, reduces the harmonic content of the current after a phase-loss fault, and improves the suppression performance of torque fluctuations after a fault. Attached Figure Description

[0055] Figure 1 This diagram illustrates the parallel drive mode of the phase-shift-free DTP-PMSM single inverter of this invention.

[0056] Figure 2 This is a control block diagram of the single inverter parallel drive phase-shift-free DTP-PMSM phase loss self-tolerance system of the present invention;

[0057] Figure 3 The diagram below shows the control block diagram of the phase-shift-free DTP-PMSM phase loss self-tolerant control system based on the improved dual-loop controller of this invention.

[0058] Figure 4(a) shows the dynamic response of the motor speed after the self-fault tolerance of the present invention (simulation);

[0059] Figure 4(b) shows the dynamic response of the motor torque after fault tolerance according to the present invention (simulation);

[0060] Figure 5(a) is a comparison analysis diagram of the phase current FFT during normal operation of the present invention (simulation);

[0061] Figure 5(b) is a comparison analysis diagram of the phase current FFT during the self-fault-tolerant operation of the present invention (simulation);

[0062] Figure 6 This is a diagram of the experimental platform built for this invention;

[0063] Figure 7(a) is a comparison of the dynamic response of motor speed after self-fault tolerance in System 1 and System 2 of the present invention;

[0064] Figure 7(b) is a comparison of the dynamic response of motor torque after self-fault tolerance in System 1 and System 2 of the present invention.

[0065] Figure 8(a) is a comparative analysis of the phase current FFT of System 1 and System 2 during normal operation of the present invention;

[0066] Figure 8(b) is a comparative analysis of the phase current FFT of System 1 and System 2 under self-fault-tolerant operation according to the present invention;

[0067] Figure 9(a) is a comparison of the dynamic response of motor speed after self-fault tolerance between System 1 of the present invention and the other systems;

[0068] Figure 9(b) is a comparison of the dynamic response of motor torque after self-fault tolerance of System 1 of the present invention and the other systems.

[0069] Figure 10(a) is a comparative analysis of the phase current FFT of System 1 of the present invention and other systems during normal operation;

[0070] Figure 10(b) is a comparative analysis of the phase current FFT of System 1 and other systems under self-fault tolerance operation according to the present invention. Detailed Implementation

[0071] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0072] A phase-loss self-tolerant fault control method for a phase-loss dual three-phase permanent magnet synchronous motor without phase shifting includes the following steps:

[0073] (I) Step 1: Two two-level three-phase voltage source inverters are connected in parallel to drive two pairs of three-phase winding modules of the phase-shiftless DTP-PMSM and a mathematical model of the three-phase permanent magnet synchronous motor is proposed.

[0074] For phase-shiftless dual three-phase permanent magnet synchronous motors (DTP-PMSMs), the traditional driving mode uses two two-level three-phase voltage source inverters to independently drive each phase of the motor, thereby achieving fault redundancy. However, this method increases the probability of motor failure and also increases the cost, size, and complexity of the driver. To address these issues, this invention proposes using two two-level three-phase voltage source inverters connected in parallel to drive the two three-phase winding modules of the phase-shiftless DTP-PMSM, such as... Figure 1 As shown, the two three-phase winding modules share the same DC bus voltage, and the line current is the sum of the phase currents from the two three-phase winding modules. Since the cost of the voltage source inverter does not increase linearly with the increase of voltage or current, the proposed drive scheme can save costs in certain situations.

[0075] The mathematical model of the electric motor is as follows:

[0076] The voltage equation in the natural reference frame is shown in equation (1):

[0077]

[0078] Among them, u jk i jk e jk R represents the voltage, current, and back electromotive force of phase j in the three-phase winding module k, respectively. sk L sk These represent the resistance and inductance of the stator in three-phase winding module k, respectively. Since this application pertains to a phase-shiftless DTP-PMSM, k = 1 or 2, and j = A, B, C or D, E, F. The parameters are considered identical between the two three-phase winding modules; therefore, R... s =R s1 =R s2 L s =L s1 =L s2 During normal operation, since the two three-phase winding modules share the same phase voltage, the corresponding phase currents of the two three-phase winding modules are the same, i.e., i A =i D i B =i E i C =i F Since each three-phase winding module operates at the same speed, the back electromotive force of the two three-phase winding modules is equal, i.e., e A =e D e B =e E e C =e F And satisfy e A +e B +e C =0.

[0079] Due to the symmetry of the motor structure, the current in each three-phase winding module is uniformly distributed, as shown in equation (2):

[0080]

[0081] Among them, i dk i qk It is the current on the dq axis; I m It is the sum of the amplitudes of the phase currents in the two three-phase winding modules. Under ideal conditions, this value is twice the amplitude of the phase current in a single three-phase winding module.

[0082] The output torque is the sum of the electromagnetic torques of the two three-phase winding modules. The control strategy is to set i... d =0. Therefore, the output torque T e It can be represented as:

[0083] T e =1.5Pψ f (iq1 +i q2 (3)

[0084] Where, ψ f The flux linkage generated by the permanent magnet magnetic field, P represents the number of pole pairs of the motor, i q1 i q2 These are the q-axis currents of three-phase winding module I and three-phase winding module II, respectively.

[0085] (II) Step 2: Automatically adjust the current distribution so that the α-axis and β-axis currents of the two pairs of three-phase winding modules are in a specific ratio; compensate the neutral point voltage and calculate the fault offset; redistribute the current to suppress the torque pulsation after the fault and eliminate the secondary ramp component caused by the phase loss fault, so as to achieve the effect of self-tolerance.

[0086] For the dual neutral point phase-shift-free DTP-PMSM used in this invention, taking the most frequent single-phase phase failure, namely A-phase phase failure, as an example, the self-tolerant phase failure scheme is derived.

[0087] Since the two neutral points are independent of each other, constraint i needs to be satisfied. A +i B +i C =0, i D +i E +i F =0. Applying Kirchhoff's voltage law, the steady-state phase current can be determined as:

[0088]

[0089] Where u β1 u β2 u α2 These are the β-axis voltages of three-phase winding module I and three-phase winding module II, and the α-axis voltage of three-phase winding module II, respectively.

[0090] The Clarke transformation matrix is:

[0091]

[0092] According to equations (4) and (5), the α-β axis current can be obtained as follows:

[0093]

[0094] Where i α1 and i α2 These represent the α-axis currents of three-phase winding module I and three-phase winding module II, respectively; i β1 and i β2 These represent the β-axis currents of three-phase winding module I and three-phase winding module II, respectively.

[0095] Since the three-phase winding module II is operating normally, and the stator winding adopts a Y-connection method, the phase voltage in this three-phase winding module is u. Dn2 =u AO- u n2O .

[0096] Since the neutral point (n1 or n2) of the motor is not connected to the DC bus voltage midpoint (O), there is no zero-sequence current, and the phase current satisfies i A +i B +i C =0 and i D +i E +i F =0, when a phase A fails, the current in A is zero, and the currents in B and C satisfy i B +i C =0. This means that the non-faulty phase currents of the faulty module exhibit the characteristics of equal amplitude, equal size, and opposite phase. After a fault occurs, the input voltage will change compared to the normal state, but it will always satisfy:

[0097]

[0098] Among them, u n1O This represents the voltage drop between the neutral point "n1" of the three-phase winding module I and the neutral point "O" of the DC bus voltage. n1O It is not 0 when the fault occurs. n1O It can be exported as:

[0099]

[0100] Since a parallel drive mode is used, the voltage relationship should satisfy u An1 =e A Therefore, the phase voltage of three-phase winding module I can be calculated by the following formula:

[0101]

[0102] Voltage can be transformed to a stationary reference frame and expressed as:

[0103]

[0104] Similarly, the phase voltage of the three-phase winding module II in the stationary coordinate system can be expressed as:

[0105]

[0106] By comparing equations (10) and (11), it can be found that when a phase-loss fault occurs in phase A, the β-axis voltages of the two three-phase winding modules are the same. Therefore, the phase current in equation (6) can be described in detail as follows:

[0107]

[0108] Among them, I α and I β These represent the current amplitudes along the α and β axes, respectively.

[0109] The Park transformation matrix is:

[0110]

[0111] According to (13), (12) can be transformed into a rotating coordinate system, as shown in formula (14):

[0112]

[0113] Substituting equation (14) into equation (3), we can obtain the electromagnetic torque of the motor during a fault as follows:

[0114]

[0115] The first term represents the average value of the output torque, while the second term represents the torque ripple. When using i... d When the control strategy is 0, the sum of the d-axis currents of three-phase winding module I and three-phase winding module II must satisfy:

[0116] i d1 +i d2 =(I α +2I β sinθcosθ≡0 (16)

[0117] It can be seen that the d-axis current in equation (16) is related to the rotor position angle. To maintain a constant zero current, I... α It must be equal to -2I β By substituting this relationship into equation (15), it should be noted that torque ripples can be completely suppressed. According to equations (3), (14), and I... α =-2I β The torque in each three-phase winding module and the total torque are:

[0118]

[0119] Among them, T e1 T is the torque of the three-phase winding module I. e2 T represents the torque of the three-phase winding module II. e The torque is the total torque of the entire motor; I βThis is the current in an orthogonal coordinate system, representing a process quantity during coordinate changes. Although the current in each three-phase winding module contains a second harmonic component, this second harmonic component in the controllable current can be significantly suppressed, and the resulting torque ripples can also be eliminated. The system control block diagram is as follows: Figure 2 As shown. Controllable current i d i q This is the sum of the dq-axis currents of the two three-phase winding modules. Due to the self-tolerant fault characteristic of the motor system in the event of a single-phase failure, no active action is required. Both the current loop and speed loop controllers of this control system use traditional PI controllers, which suffer from poor dynamic response and robustness, thus requiring further improvements to the controllers.

[0120] (III) Step 3: Improve the current loop controller by adopting deadbeat predictive control.

[0121] The current loop controller in the control system is improved with the goal of reducing current harmonic components. A deadbeat predictive current controller (DPCC) is used instead of the traditional PI controller. The continuous voltage equation is discretized and the ideal control voltage is calculated based on the current reference value at the next moment, so as to achieve high dynamic response and reduce harmonic distortion.

[0122] The state equation of the surface-mounted, phase-shiftless DTP-PMSM in the dp rotating coordinate system is as follows:

[0123]

[0124] Where u d u q The voltage component along the dq axis; L d L q The stator is equivalent to the inductance on the dq axis; ψ f For permanent magnet flux linkage; ω e The electric angle is in radians.

[0125] Because the sampling time T s Since the equation is short enough, the first-order Euler forward discretization method is used to discretize the above equation, resulting in:

[0126]

[0127] Substituting equation (19) into equation (18) and rewriting it in the form of a matrix equation:

[0128]

[0129] Based on the fundamental principle of deadbeat current prediction, predicting the desired voltage requires using the current given value as the current value at the next moment, i.e.:

[0130]

[0131] Combining equations (20) and (21), the ideal deadbeat control voltage is obtained as follows:

[0132]

[0133] Among them, u d * (k) represents the ideal control voltage along the d-axis at time k, calculated using a predictive model; u q * (k) represents the ideal control voltage along the q-axis at time k, calculated using a predictive model; i d * (k+1) is the reference value of the d-axis current at time k+1, given by the outer loop controller; i q * (k+1) is the reference value of the q-axis current at time k+1, given by the outer loop controller; ω e (k) is the electric angular velocity of the motor at time k, which is the core parameter for calculating the back electromotive force and coupling terms.

[0134] (iv) Step 4: Speed ​​loop controller optimization design to improve the sliding film controller.

[0135] A novel sliding mode surface incorporating rotational speed error integrals and nonlinear functions is constructed, and the sigmoid function is used to optimize the reaching law to suppress chattering. An integrated disturbance observer is used to construct a second-order model to estimate disturbances, and a high-gain feedback mechanism is added to compensate for load abrupt changes, thereby improving the system's dynamic response capability. The specific details are as follows:

[0136] The mathematical model for a single inverter driving a motor in parallel is as follows:

[0137]

[0138] Where J is the moment of inertia, T L This represents the load torque.

[0139] Define the state variables of a phase-shift-free DTP-PMSM system:

[0140] Where, ω ref The reference speed for the motor is ω, which is usually a constant. m This represents the actual speed of the motor. Combining equations (23) and (24), we can obtain:

[0141]

[0142] definition M = 3Pψ f / 2J, then formula (25) can be transformed into:

[0143]

[0144] The traditional definition of the slid surface function for slid control is:

[0145]

[0146] Where c>0 represents the parameter to be designed.

[0147] Differentiating formula (27), we get:

[0148]

[0149] Traditional synovial control generally uses an exponential reaching law, which yields:

[0150]

[0151] Therefore, the reference current along the q-axis can be obtained as:

[0152]

[0153] As can be seen from the traditional exponential reaching law, the presence of a discontinuous term -εsgn(s) in the reaching law will cause the system to chatter; the larger the value of ε, the stronger the robustness and the faster the reaching rate, but the greater the chattering will also be.

[0154] To achieve stable system operation, the sigmoid function is used instead of the sign function. A novel exponential reaching law and sliding surface function based on the sigmoid function are proposed to suppress the inherent chattering of the sliding mode.

[0155] Define the sliding surface function as:

[0156]

[0157] The new exponential reaching law based on the sigmoid function is as follows:

[0158]

[0159] Therefore, the reference current for the q-axis after the improved sliding film control can be obtained as follows:

[0160]

[0161] Among them, T L ω represents the load torque, characterizing the external resistance at the motor shaft end; J represents the rotor, reflecting the inertia of the mechanical system; mLet be the mechanical angular velocity, defining the actual output speed; k1 and k2 are the sliding surface control gains, adjusting the linear and nonlinear error compensation strengths respectively; M is the torque constant, establishing the current-torque conversion relationship; B is the viscous friction coefficient; x1 is the state variable, representing the speed tracking error; σ(s) is the switching component derived from the sigmoid function, replacing the traditional sign function to suppress chattering σ(s) = sigmoid(s); ε2 and q2 are the reaching law control parameters, controlling the disturbance rejection strength and convergence speed respectively. The sigmoid function in the formula is replaced by its abbreviation σ to prove the stability of the designed controller.

[0162] Select the Lyapunov function:

[0163]

[0164] Differentiating and substituting into equation (32), we get:

[0165]

[0166] From equation (23), considering the changes in system parameters and load torque, we can obtain:

[0167]

[0168] In the formula, a = -F / J, b = 3p n ψ f / 2J; u is the control law to be designed; Δa, Δb, Δc represent the parameter changes of the motor; d represents the total system disturbance, including internal parameter disturbance and external load disturbance.

[0169] Combining (34) and (37), the designed control law is:

[0170]

[0171] As can be seen from equation (38), the existence of the uncertainty term d has a significant impact on the control performance of the system.

[0172] Therefore, let the state variable z1 = ω m z2 = -d, output y = ω m Because the disturbance d in a real motor system changes slowly, its first derivative can be considered as zero. Therefore, the state-space equation of the system is established as follows:

[0173]

[0174] Using z1 and z2 as the observation objects, a high-gain feedback mechanism for the rotational speed estimation error e is established, and the disturbance observer is designed as follows:

[0175]

[0176] In the formula, z1 is the actual input mechanical angular velocity; The mechanical angular velocity estimated by the observer; For the estimated matching perturbation, α1 and α2 are positive real numbers. This perturbation observer can then be used to achieve...

[0177] As shown in Equation (40), the load disturbance observed by the disturbance observer is compensated into the speed loop controller to improve the system's anti-interference performance.

[0178]

[0179] The novel speed loop controller designed as shown in equation (41) is called the Improved Sliding Film Controller (ISMC).

[0180] (V) Step 5: Combine the improved sliding mode control and deadbeat predictive control to form an improved dual-loop controller system. Set high gain parameters to improve the dynamic response performance of the system and form a new self-fault-tolerant control system.

[0181] The improved sliding diaphragm controller (ISMC) designed according to equation (41) replaces the traditional sliding diaphragm controller (SMC) used in the speed loop, and the deadbeat predictive current controller (DPCC) designed according to equation (12) replaces the traditional current loop PI controller. The improved system control block diagram is as follows. Figure 3 As shown.

[0182] To verify the effectiveness of this invention, simulation verification and comparison were performed using Simulink. The motor parameters used in the simulation are shown in Table 1, and the speed loop PI controller parameters are: k pω =0.4, k iω =15, the current loop PI controller parameters are: k pd =k pq =1.2, k id =k iq =28.88, sampling time T s =1×10 -6 The SMC controller parameters are: c = 60, ε1 = 200, q1 = 300. The ISMC controller parameters are: k1 = 0.001, k2 = 0.01, ε2 = 18, q2 = 250, α1 = 5, α2 = 20, a0 = 10.

[0183] Table 1 Parameters of Phase-Shift-Free Dual Three-Phase Permanent Magnet Synchronous Motors

[0184]

[0185] Given a rated speed of 2000 rpm, using i d=0 control, with an initial load torque of 0.15 N·m. It is set that a phase loss fault occurs in phase A at 0.2s. After entering the self-fault-tolerant state and stabilizing, the load torque suddenly changes from 0.15 N·m to 0.25 N·m at 0.3s. The improved control and the traditional control are compared with the data. The simulation results are shown in Figures 4 and 5.

[0186] As shown in Figure 4, the dynamic response of the motor after fault tolerance reveals that the traditional PI controller-based fault tolerance control exhibits significant speed fluctuations and torque overshoot after a sudden torque change in the fault-tolerant state, with a relatively long time to reach stability. In contrast, the improved dual-loop controller significantly reduces speed fluctuations and torque overshoot after a sudden torque change, and the time to stability is greatly shortened, preliminarily verifying the superiority of the improved control system in dynamic response performance. Figure 5, comparing the harmonic content of each phase current, shows that under the same conditions, the improved dual-loop controller reduces the harmonic content of each phase current compared to the PI controller, both during normal operation and after a phase loss during fault tolerance operation, preliminarily verifying the superiority of the improved control system in fault tolerance performance.

[0187] To further verify the effectiveness of the proposed method, a system was built. Figure 6 The experimental platform shown is identical to the simulation parameters. The motor parameters used in the experiment are consistent with the simulation parameters, as are the controller parameters. The parameter settings for each operating condition are also consistent with the simulation, and i... d =0 control. To verify the rationality and effectiveness of the proposed algorithm, five phase-shift-free DTP-PMSM self-tolerant control systems were built on this experimental platform. System 1: Both the speed loop and current loop use PI controllers; System 2: The speed loop uses a PI controller, and the current loop uses a DPCC controller; System 3: The speed loop uses a traditional exponentially reaching SMC controller, and the current loop uses a PI controller; System 4: The speed loop uses the ISMC designed in this paper, and the current loop is the same as System 3; System 5: The speed loop uses the ISMC designed in this paper, and the current loop uses a DPCC controller (the parameters of the same controller used in each system are kept consistent).

[0188] The self-fault-tolerant performance and dynamic response performance of System 1 and System 2 after a phase loss fault were compared. The speed was 2000 rpm, the load was 0.15 N·m, and after stable operation, a phase loss fault was induced in phase A via a circuit breaker. The experimental results before and after the fault are shown in Figures 7 and 8. The results show that, compared to the traditional PI controller, replacing the current loop controller with a DPCC controller has little impact on the dynamic response of the motor speed and torque, but it significantly improves the suppression of torque fluctuations and phase current harmonics after a fault.

[0189] The self-fault-tolerant performance of System 1 and the other three systems after a phase loss fault, as well as the dynamic response performance of the system under self-fault-tolerant conditions, were compared. At a speed of 2000 rpm and a load of 0.15 N·m, after stable operation, a phase loss fault was induced in phase A via a circuit breaker. After stabilization, a sudden torque change from 0.15 N·m to 0.25 N·m was set. The experimental comparison results are shown in Figures 9 and 10. The experimental effects of each system under different operating conditions are compared in Table 2.

[0190] Table 2 Comparison of experimental results for each system under different operating conditions

[0191]

[0192] Through theoretical analysis, simulation verification, and experimental research, compared with the traditional phase-loss self-tolerance control strategy, the phase-loss self-tolerance control strategy based on the improved dual-loop controller proposed in this paper can significantly improve the anti-interference performance and dynamic response performance of the motor system, reduce the harmonic content of the current after a phase-loss fault, and improve the suppression of torque fluctuations after the fault. This further reduces the negative impact of phase-loss faults on the system and improves the self-tolerance of the motor system after a phase-loss fault. It provides a new idea and method for phase-loss fault-tolerant control of phase-loss-free DTP-PMSM.

[0193] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A phase-loss self-tolerant fault control method for a phase-loss dual three-phase permanent magnet synchronous motor without phase shift, characterized in that, Includes the following steps: Step 1: Two two-level three-phase voltage source inverters are connected in parallel to drive two pairs of three-phase winding modules of a phase-shiftless DTP-PMSM and a mathematical model of a three-phase permanent magnet synchronous motor is proposed. Step 2: Automatically adjust the current distribution to make the α-axis and β-axis currents of the two pairs of three-phase winding modules in a specific ratio; compensate for the neutral point voltage and calculate the offset of the faulty module; The current is redistributed to suppress torque pulsation after a fault and eliminate the secondary ramp component caused by the phase loss fault, so as to achieve the effect of self-fault tolerance. Step 3: Improve the current loop controller by adopting deadbeat predictive control; Step 4: The speed loop controller is optimized and redesigned as an improved sliding diaphragm controller; Step 5: Combine improved sliding mode control and deadbeat predictive control to form an improved dual-loop controller system. Set high-gain parameters to improve the dynamic response performance of the system and form a new self-fault-tolerant control system.

2. The self-fault-tolerant phase-loss control method for a phase-shiftless dual three-phase permanent magnet synchronous motor according to claim 1, characterized in that, In step 1, the two pairs of three-phase windings share the same DC bus voltage, and the line current is the sum of the phase currents from the two pairs of three-phase windings; therefore, the mathematical model of the motor is as follows: Among them, i dk i qk It is the current on the dq axis, k = 1 or 2; I m It is the sum of the amplitudes of the phase currents in the two pairs of three-phase winding modules; the output torque of the three-phase permanent magnet synchronous motor is the sum of the electromagnetic torques of the two pairs of three-phase winding modules, i d =0, output torque T e It can be represented as: T e =1.5Pψ f (i q1 +i q2 ) (3) Where, ψ f The flux linkage generated by the permanent magnet magnetic field, P represents the number of pole pairs of the motor, i q1 i q2 These are the q-axis currents of three-phase winding module I and three-phase winding module II, respectively.

3. The self-fault-tolerant phase-loss control method for a phase-shiftless dual three-phase permanent magnet synchronous motor according to claim 2, characterized in that, In step 2, the torque in each three-phase winding module and the torque of the entire motor are: Among them, T e1 T is the torque of the three-phase winding module I. e2 T represents the torque of the three-phase winding module II. e The torque is the total torque of the entire motor; I β It is the current in an orthogonal coordinate system, and it is a process quantity of coordinate change.

4. The self-fault-tolerant phase-loss control method for a phase-shiftless dual three-phase permanent magnet synchronous motor according to claim 2, characterized in that, In step 3, a deadbeat predictive current controller is used to replace the traditional PI controller, the continuous voltage equation is discretized, and the ideal control voltage is calculated based on the current reference value at the next moment, so as to achieve high dynamic response and reduce harmonic distortion.

5. The self-fault-tolerant phase-loss control method for a phase-shiftless dual three-phase permanent magnet synchronous motor according to claim 4, characterized in that, The state equations of the surface-mounted, phase-shiftless, dual three-phase permanent magnet synchronous motor in the dp rotating coordinate system are as follows: Among them, u d u q The voltage component along the dq axis; L d L q The stator is equivalent to the inductance on the dq axis; ψ f For permanent magnet flux linkage; ω e The electric angle is in radians. Because the sampling time T s Since the equation is short enough, the first-order Euler forward discretization method is used to discretize the above equation, resulting in: Substituting equation (19) into equation (18) and rewriting it in the form of a matrix equation: Based on the fundamental principle of deadbeat current prediction, predicting the desired voltage requires using the current given value as the current value at the next moment, i.e.: Combining equations (20) and (21), the ideal deadbeat control voltage is obtained as follows: Among them, u d * (k) represents the ideal control voltage along the d-axis at time k, calculated using a predictive model; u q * (k) represents the ideal control voltage along the q-axis at time k, calculated using a predictive model; i d * (k+1) is the reference value of the d-axis current at time k+1, given by the outer loop controller; i q * (k+1) is the reference value of the q-axis current at time k+1, given by the outer loop controller; ω e (k) is the electric angular velocity of the motor at time k, which is the core parameter for calculating the back electromotive force and coupling terms.

6. The self-fault-tolerant phase-loss control method for a phase-shiftless dual three-phase permanent magnet synchronous motor according to claim 1, characterized in that, In step 4, an improved sliding mode control is designed, a novel sliding mode surface containing speed error integral and nonlinear function is constructed, and the sigmoid function is used to optimize the reaching law to suppress chattering; and a disturbance observer is integrated to construct a second-order model to estimate disturbances, and a high-gain feedback mechanism is added to compensate for load mutations, thereby improving the dynamic response capability of the system.

7. The self-fault-tolerant phase-loss control method for a phase-shiftless dual three-phase permanent magnet synchronous motor according to claim 6, characterized in that, In step 4, the sliding surface function is defined as: The new exponential reaching law based on the sigmoid function is as follows: Therefore, the reference current for the q-axis after the improved sliding film control can be obtained as follows: Among them, T L ω represents the load torque, characterizing the external resistance at the motor shaft end; J represents the rotor, reflecting the inertia of the mechanical system; m ε is the mechanical angular velocity, defining the actual output speed; k1 and k2 are the sliding surface control gains, which adjust the linear and nonlinear error compensation strengths respectively; M is the torque constant, establishing the current-torque conversion relationship; B is the viscous friction coefficient; x1 is the state variable, representing the speed tracking error; σ(s) is the switching component derived from sigmoid, replacing the traditional sign function to suppress chattering σ(s) = sigmoid(s); ε2 and q2 are the reaching law control parameters, controlling the disturbance rejection strength and convergence speed respectively.

8. The self-fault-tolerant phase-loss control method for a phase-shiftless dual three-phase permanent magnet synchronous motor according to claim 7, characterized in that, The control law for the motor is: Let the state variable z1 = ω m z2 = -d, output y = ω m Since the disturbance d in the actual motor system changes slowly, its first derivative can be considered as 0. Therefore, the state-space equation of the system is established as follows: Using z1 and z2 as the observation objects, a high-gain feedback mechanism for the rotational speed estimation error e is established, and the disturbance observer is designed as follows: In the formula, z1 is the actual input mechanical angular velocity; The mechanical angular velocity estimated by the observer; For the estimated matching perturbation, α1 and α2 are positive real numbers; through this perturbation observer, it is possible to achieve a = -F / J, b = 3p n ψ f / 2J; u represents the control law to be designed; As shown in equation (40), the load disturbance observed by the disturbance observer is compensated into the speed loop controller, thus yielding: The novel speed ring controller designed as shown in equation (41) is called the improved sliding film controller.

Images