Stator winding ripple current modeling and predicting method suitable for fault asymmetric multi-phase motor system

Abstract

The invention discloses a stator winding ripple current modeling and predicting method suitable for a fault asymmetric multi-phase motor system, and the method comprises the following steps: constructing an inductance matrix of a motor based on the self-inductance and inter-phase mutual inductance of each phase winding of the motor; performing elementary transformation on a motor voltage equation under a natural coordinate system based on the inductance matrix and a preset orthogonal matrix to obtain a motor voltage equation under a decoupling coordinate system, and obtaining an expression of a winding current slope based on an instantaneous voltage equation under a bridge arm open-circuit fault; on the basis of a motor voltage equation under a decoupling coordinate system and a forced constraint that a fault phase current is zero, representing a neutral point voltage of the motor as a weighted sum of a residual healthy bridge arm output voltage and mutual inductance and a rotor flux differential item; on the basis of the weighted sum expression of the output voltage and the neutral point voltage of the remaining healthy bridge arms, the instantaneous phase voltage of each phase winding is obtained through calculation; and based on the expression of the instantaneous phase voltage and the winding current slope, the ripple current of the motor is obtained through integral calculation.

Description

Technical Field

[0001] This invention belongs to the field of motor drive technology, and in particular relates to a method for modeling and predicting stator winding ripple current applicable to faulty asymmetrical multiphase motor systems. Background Technology

[0002] Variable frequency drive systems, as the core of DC-AC power conversion equipment, are widely used in aerospace, electric vehicles, and ship propulsion. To achieve flexible speed regulation, pulse width modulation (PWM) technology based on the volt-second equivalent principle is typically used to control power switching devices, outputting a high-frequency voltage pulse sequence to equivalently generate the required AC voltage. However, while PWM technology reduces harmonic losses and improves the dynamic response performance of the system, the discrete high-frequency pulse voltage generated by PWM chopping produces a significant switching current ripple in the motor stator windings. Since the switching current ripple directly affects key performance characteristics of the motor system, such as torque pulsation, vibration noise, and differential-mode electromagnetic interference, accurate prediction of it becomes a crucial foundation for system optimization design.

[0003] Furthermore, the switching devices in frequency converters operate under harsh environments with multiple high loads for extended periods, switching on and off tens of thousands of times per second under these conditions. This makes them the most vulnerable component in the system, thus making the frequency converter one of the weakest links. Therefore, improving the fault tolerance of the frequency converter drive system is crucial to ensuring its long-term, efficient, safe, and stable operation.

[0004] Multiphase electric drive systems, with their numerous controlled arms and windings and redundant backups between phases, provide a hardware foundation for fault-tolerant control and are therefore widely used in applications requiring high safety and reliability. However, further improvements in the fault-tolerant performance of multiphase motor systems face two challenges: First, after a motor system failure, the inverter enters an asymmetrical operating state, leading to more pronounced inter-phase coupling and significant changes in current ripple characteristics, thus significantly increasing the complexity of ripple prediction. Second, most existing current ripple prediction methods are based on the premise that "the voltage at the arm output terminal to ground can be linearly represented by a switching function." However, under fault conditions, the arm port voltage becomes uncontrollable, rendering this premise invalid. Existing ripple current prediction models and methods for healthy symmetrical electric drive systems are difficult to apply, thus limiting performance optimization under fault-tolerant operation. Summary of the Invention

[0005] To address the aforementioned technical problems, this invention proposes a method for modeling and predicting stator winding ripple current in fault-prone multiphase motor systems, thereby resolving the issues present in the prior art.

[0006] To achieve the above objectives, this invention provides a method for modeling and predicting stator winding ripple current in fault-prone multiphase motor systems, comprising: Based on the self-inductance of each phase winding and the mutual inductance between phases, the inductance matrix of the motor is constructed. Based on the inductance matrix and the preset orthogonal matrix, the motor voltage equation in the natural coordinate system is transformed into the motor voltage equation in the decoupled coordinate system, and the expression for the winding current slope is obtained based on the instantaneous voltage equation under the bridge arm open circuit fault. Based on the motor voltage equation in the decoupled coordinate system and the forced constraint that the fault phase current is zero, the motor neutral point voltage is characterized as a weighted sum of the output voltage of the remaining healthy bridge arm and the differential term of the rotor flux linkage. The instantaneous phase voltage of each phase winding is calculated based on the weighted sum expression of the output voltage of the remaining healthy bridge arm and the neutral point voltage. Based on the expressions for the instantaneous phase voltage and the winding current slope, the ripple current of the motor is calculated by integration.

[0007] Optionally, the windings of each phase of the motor can be processed using finite element simulation or experimental measurement to obtain the self-inductance and inter-phase mutual inductance of each phase winding of the motor.

[0008] Optionally, the motor voltage equation in the decoupled coordinate system is: ; In the formula, [ u d ], [ L d ], [ i d ], [ ψ fd [These represent the voltage, inductance, current, and rotor flux linkage matrices in the decoupled coordinate system after orthogonal matrix mapping.]

[0009] Optionally, the orthogonal matrix can achieve similar diagonalization of the inductance matrix. When the motor windings are distributed as sinusoidal windings, the orthogonal matrix is ​​a Clarke transformation matrix.

[0010] Optionally, the process of characterizing the motor neutral point voltage as a weighted sum of the output voltage of the remaining healthy bridge arm and the differential term of the rotor flux linkage includes: Substituting the constraint that the fault phase current is zero and its differential is zero into the motor voltage equation in the decoupled coordinate system, we obtain the first relationship including the neutral point voltage. The inverse matrix of the orthogonal matrix is ​​used to transform the first relation back to the natural coordinate system to obtain the second relation between the neutral point voltage, the output voltage of the remaining healthy bridge arm, and the differential term of the rotor flux linkage. The weighted sum is obtained based on the first relation and the second relation.

[0011] Optionally, the process of calculating the instantaneous phase voltage of each phase winding includes: The switching cycle of a single switching cycle is divided into several segments by using the switching action time of the remaining healthy bridge arm; For each segment, the instantaneous phase voltage of each phase winding in the current segment is calculated based on the output voltage of the remaining healthy bridge arm in the current segment and combined with the weighted sum expression of the neutral point voltage.

[0012] Optionally, the process of calculating the ripple current of the motor through integration includes: The instantaneous phase voltage is processed using the orthogonal matrix and mapped to the decoupled coordinate system to obtain the instantaneous phase voltage in the decoupled coordinate system; Based on the expressions for the instantaneous phase voltage and the winding current slope in the decoupled coordinate system, the current ripple slope in the decoupled coordinate system is obtained. The inverse matrix of the orthogonal matrix is ​​used to process the current ripple slope in the decoupled coordinate system, and then transformed to the natural coordinate system to obtain the current ripple slope in the natural coordinate system. The current ripple slope in the natural coordinate system is integrated in each segment, and the initial value of the integration is corrected based on the current sampling value at the beginning of each switching cycle to obtain the final ripple current.

[0013] Compared with the prior art, the present invention has the following advantages and technical effects: This invention establishes a quantitative analysis model for current ripple under bridge arm open-circuit faults by constructing an inductance matrix and applying orthogonal transformations to decouple the voltage equations. This method comprehensively considers the influence of high- and low-frequency flux linkage components on ripple, and derives the precise relationship between the neutral point voltage and the healthy phase output voltage using the constraint that the fault phase current is zero, thus achieving accurate prediction of current ripple in fault-prone asymmetric systems. Compared to existing methods, this invention does not require complex reduced-order decoupling matrix design, exhibits good adaptability to motors with any number of phases, and provides a reliable theoretical foundation and technical support for ripple suppression and system optimization under fault-tolerant operation. Attached Figure Description

[0014] The accompanying drawings, which form part of this application, are used to provide a further understanding of this application. The illustrative embodiments and descriptions of this application are used to explain this application and do not constitute an undue limitation of this application. In the drawings: Figure 1 This is a flowchart of a method according to an embodiment of the present invention; Figure 2 This is a schematic diagram of the equivalent circuit of a single-phase winding of a motor according to an embodiment of the present invention; Figure 3 This is a schematic diagram of the ripple current segmentation within a single switching cycle according to an embodiment of the present invention; Figure 4 The following is an overall illustration comparing the simulation results of the first phase bridge arm open circuit fault and the current ripple of the fifth phase winding of the motor under SPWM modulation, as well as the prediction calculation results without considering low-frequency flux linkage and with consideration of low-frequency flux linkage, in accordance with the embodiments of the present invention. Figure 5 This is a magnified view showing the simulation of the first phase bridge arm open circuit fault and the current ripple of the fifth phase winding of the motor under SPWM modulation, as well as the comparison of the prediction calculation results without considering low-frequency flux linkage and with consideration of low-frequency flux linkage in an embodiment of the present invention. Detailed Implementation

[0015] It should be noted that, unless otherwise specified, the embodiments and features described in this application can be combined with each other. This application will now be described in detail with reference to the accompanying drawings and embodiments.

[0016] It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions, and although a logical order is shown in the flowchart, in some cases the steps shown or described may be executed in a different order than that shown here.

[0017] Example 1 like Figure 1 As shown, this embodiment provides a method for modeling and predicting stator winding ripple current applicable to fault-prone multiphase motor systems. The common-mode voltage analysis and suppression method proposed in this invention is also applicable to motor systems with any number of phases experiencing simultaneous open-circuit faults in multiphase bridge arms. The method includes the following steps: (1) The inductance matrix of the motor is constructed by measuring the self-inductance and inter-phase mutual inductance of each phase winding through finite element simulation or experiment. In a specific example, a five-phase permanent magnet synchronous motor system is used as an example for illustration, and its inductance matrix is ​​shown in Table 1.

[0018] Table 1 (2) Apply orthogonal matrices to perform elementary transformations on the motor equations to realize the mapping of the motor voltage equations from the phase-coupled natural coordinate system to the phase-decoupled decoupled coordinate system. At the same time, in the decoupled coordinate system, based on the instantaneous voltage equations under the bridge arm open circuit fault of the multiphase motor system, calculate the expression of the winding current slope.

[0019] The proposed calculation of winding ripple current slope mainly includes the following steps: a. The voltage equations of the motor are decoupled and mapped using an orthogonal matrix [T]. The orthogonal matrix [T] is an m×m matrix, where m is the number of phases in the motor. The orthogonal matrix [T] should be able to diagonalize the inductance matrix. Physically, this means that elementary transformations are used to map the flux linkages generated by the mutual inductance of each phase winding and the self-inductance of a single-phase winding to the equivalent flux linkages generated by the self-inductance of a single phase, thus achieving decoupling of the phase windings. For example... Figure 2 As shown. The specific implementation process includes: Elementary transformations are performed on the motor voltage equations to ensure that the equivalent single-phase current of the motor is only affected by the equivalent single-phase voltage, flux linkage, and inductance. These elementary transformations are then equivalent to left-multiplying the motor voltage equations by matrix [T]. Where [T] represents an orthogonal matrix that enables the diagonalization of the inductance matrix, m is the number of motor phases, and t ij With s ij These are the elements of an orthogonal matrix and its inverse: [u], [R], [i], [L] s ]、[ψ f These represent the phase voltage, phase resistance, phase current, inductance, and rotor flux linkage matrix of the motor system in the natural coordinate system. θ Represents the rotor position electrical angle, ψ f This represents the amplitude of the rotor flux linkage. It should be noted that the low-frequency harmonics generated by the rotor flux linkage are ignored here, and the back electromotive force of the rotor flux linkage excitation is assumed to be a sine wave.

[0020] In a specific example, simply substitute m=5.

[0021] b. Solve for the ripple current slope expression in the decoupled coordinate system. Considering that the voltage drop across the motor stator resistance is usually negligible, the voltage equation in the decoupled coordinate system can be expressed as: in,[ u d ], [ L d ], [ i d ], [ ψ fdThese are the voltage, inductance, current, and rotor flux linkage matrices in the decoupled coordinate system after orthogonal matrix mapping: At this point, the phase voltage of the motor is entirely determined by the differential terms of the magnetic flux generated by the stator and rotor. Therefore, the current slope in the decoupled coordinate system can be directly solved: (3) Based on the motor voltage equation in the decoupled coordinate system, the forced constraint that the fault phase current is 0 is substituted, and the low and high frequency components in the equation are considered at the same time. The neutral point voltage of the motor is expressed as the weighted sum of the output voltage of the remaining healthy bridge arm and the differential term of the rotor flux linkage.

[0022] The proposed method for calculating the neutral point voltage weighting coefficient mainly includes the following steps: a. Considering the constraint that the fault phase current is forced to be 0, the derivative of its phase current also remains 0. In a specific example, we will illustrate this with an open-circuit fault in the first phase arm: It should be emphasized that the differential term of the low-frequency rotor flux here cannot be ignored.

[0023] b. Inversely transform the phase voltage and rotor flux in the decoupled coordinate system of the equations obtained in step a to the natural coordinate system: Since an open-circuit fault in a bridge arm can be equivalent to a healthy motor system being subjected to a forced constraint that the fault phase current is zero, the neutral point voltage of the motor can be expressed as: c. By simultaneously solving the above equations, the neutral point voltage of the motor can be expressed as a weighted sum of the output voltage of the remaining healthy bridge arm and the differential term of the rotor flux linkage: (4) Substitute the output voltage of the remaining healthy bridge arm into the neutral point voltage expression obtained in step S3 to calculate the instantaneous phase voltage of each phase winding. Divide the single switching cycle according to the switching time of the remaining healthy bridge arm. In a specific example, after a single-phase bridge arm open-circuit fault occurs in the five-phase permanent magnet synchronous motor, there are a total of 8 switching transients in the remaining four phase bridge arms. Therefore, divide each switching cycle into 9 segments according to the switching time from smallest to largest, and calculate the phase voltage magnitude within each time window, such as... Figure 3As shown. Furthermore, since SPWM modulation is used in this specific example, the pulse voltages of each phase are mirror-symmetrical within one switching cycle, so only the first five segments need to be calculated.

[0024] (5) The instantaneous phase voltage values ​​of each phase winding are mapped to the decoupled coordinate system through the orthogonal matrix [T] for solution, and then solved by the inverse matrix [T] of the orthogonal matrix. -1 The inverse transformation is performed to the natural coordinate system, and integration is performed within each sub-time window to obtain the instantaneous values ​​of the winding phase currents: It is important to note that, considering the unavoidable prediction errors caused by calculation errors, dead time, and parameter errors in practical applications, the initial value of the predicted current in each switching cycle should be corrected to the current sampling value to minimize prediction errors. In a specific example, taking an open-circuit fault in the first phase arm as an example, the prediction method proposed in this invention is applied to predict the ripple current of the fifth phase winding. The simulation results of the motor's fifth phase winding current ripple under SPWM modulation, and the comparison of prediction calculation results without considering low-frequency flux linkage with those considering low-frequency flux linkage, are shown below. Figure 4 and Figure 5 As shown, it can be seen that since the influence of low-frequency flux on ripple current is difficult to ignore, the neglect of low-frequency flux in existing methods will lead to significant prediction errors. However, the modeling and ripple current prediction method proposed in this invention can more accurately predict the instantaneous magnitude of the winding current.

[0025] In summary, the ripple current modeling and prediction method proposed in this invention can achieve accurate prediction of ripple current in multiphase motor systems under open-circuit faults in bridge arms.

[0026] This invention addresses the shortcomings of existing ripple prediction models that fail under open-circuit faults in bridge arms. Based on the circuit modal characteristics of the electric drive system, it remodels the phase voltage and current ripple expressions of the motor stator winding under asymmetrical power supply faults, achieving a quantitative characterization of the output current ripple. This makes the ripple formation mechanism and prediction analysis of the motor system during fault-tolerant operation clearer, which is beneficial for evaluating the output quality and steady-state performance of the system under all operating conditions. It also provides a basis for the ripple suppression and fault-tolerant control design of high-reliability motor drive systems.

[0027] Compared with existing modeling methods for faulty electric drive systems, this invention simultaneously calculates the impact of high- and low-frequency flux linkage components on current ripple, comprehensively considers the contribution of each frequency band component to current ripple, and significantly improves prediction accuracy.

[0028] This invention starts with the analysis of motor voltage equations in the natural coordinate system. Through elementary matrix transformations and the forced electrical constraints introduced by circuit faults, it analyzes the motor model under fault-tolerant operation of the electric drive system. Compared with existing analysis methods based on reduced-order decoupling matrices, it does not require the design of complex reduced-order decoupling matrices and has better adaptability to different motor phase numbers.

[0029] The above are merely preferred embodiments of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A method for modeling and predicting stator winding ripple current in a fault-prone multiphase motor system, characterized in that, include: Based on the self-inductance of each phase winding and the mutual inductance between phases, the inductance matrix of the motor is constructed. Based on the inductance matrix and the preset orthogonal matrix, the motor voltage equation in the natural coordinate system is transformed into the motor voltage equation in the decoupled coordinate system, and the expression for the winding current slope is obtained based on the instantaneous voltage equation under the bridge arm open circuit fault. Based on the motor voltage equation in the decoupled coordinate system and the forced constraint that the fault phase current is zero, the motor neutral point voltage is characterized as a weighted sum of the output voltage of the remaining healthy bridge arm and the differential term of the rotor flux linkage. The instantaneous phase voltage of each phase winding is calculated based on the weighted sum expression of the output voltage of the remaining healthy bridge arm and the neutral point voltage. Based on the expressions for the instantaneous phase voltage and the winding current slope, the ripple current of the motor is calculated by integration.

2. The method for modeling and predicting stator winding ripple current in a fault-prone multiphase motor system according to claim 1, characterized in that, The self-inductance and inter-phase mutual inductance of each phase winding of the motor are obtained by processing the windings of each phase using finite element simulation or experimental measurement.

3. The method for modeling and predicting stator winding ripple current in a fault-prone multiphase motor system according to claim 1, characterized in that, The motor voltage equation in the decoupled coordinate system is: ； In the formula, [ u d ], [ L d ], [ i d ], [ ψ fd [These represent the voltage, inductance, current, and rotor flux linkage matrices in the decoupled coordinate system after orthogonal matrix mapping.] 4. The method for modeling and predicting stator winding ripple current in a fault-prone multiphase motor system according to claim 1, characterized in that, The orthogonal matrix can achieve similar diagonalization of the inductance matrix. When the motor windings are distributed as sinusoidal windings, the orthogonal matrix is ​​a Clarke transformation matrix.

5. The method for modeling and predicting stator winding ripple current in a fault-prone multiphase motor system according to claim 1, characterized in that, The process of characterizing the neutral point voltage of the motor as a weighted sum of the output voltage of the remaining healthy bridge arm and the differential term of the rotor flux linkage includes: Substituting the constraint that the fault phase current is zero and its differential is zero into the motor voltage equation in the decoupled coordinate system, we obtain the first relationship including the neutral point voltage. The inverse matrix of the orthogonal matrix is ​​used to transform the first relation back to the natural coordinate system to obtain the second relation between the neutral point voltage, the output voltage of the remaining healthy bridge arm, and the differential term of the rotor flux linkage. The weighted sum is obtained based on the first relation and the second relation.

6. The method for modeling and predicting stator winding ripple current in a fault-prone multiphase motor system according to claim 1, characterized in that, The process of calculating the instantaneous phase voltage of each phase winding includes: The switching cycle of a single switching cycle is divided into several segments by using the switching action time of the remaining healthy bridge arm; For each segment, the instantaneous phase voltage of each phase winding in the current segment is calculated based on the output voltage of the remaining healthy bridge arm in the current segment and combined with the weighted sum expression of the neutral point voltage.

7. The method for modeling and predicting stator winding ripple current in a fault-prone multiphase motor system according to claim 1, characterized in that, The process of calculating the ripple current of the motor through integration includes: The instantaneous phase voltage is processed using the orthogonal matrix and mapped to the decoupled coordinate system to obtain the instantaneous phase voltage in the decoupled coordinate system; Based on the expressions for the instantaneous phase voltage and the winding current slope in the decoupled coordinate system, the current ripple slope in the decoupled coordinate system is obtained. The inverse matrix of the orthogonal matrix is ​​used to process the current ripple slope in the decoupled coordinate system, and then transformed to the natural coordinate system to obtain the current ripple slope in the natural coordinate system. The current ripple slope in the natural coordinate system is integrated in each segment, and the initial value of the integration is corrected based on the current sampling value at the beginning of each switching cycle to obtain the final ripple current.

Images