A hybrid signal injection-based online parameter identification method for interior permanent magnet synchronous motors

Abstract

The application discloses an online parameter identification method of an embedded permanent magnet synchronous motor with mixed signal injection. Firstly, a mathematical model of the embedded permanent magnet synchronous motor considering the difference between the fundamental frequency resistance and the high frequency resistance is provided, including a steady-state incremental model of the fundamental frequency and a small signal dynamic model of the high frequency. Secondly, according to the high frequency small signal dynamic model of the embedded permanent magnet synchronous motor, an online parameter identification method based on double-frequency sinusoidal current signal injection is provided, which is used for calculating the incremental self-induction and the incremental mutual-induction under the operating condition point. Finally, according to the steady-state incremental model of the fundamental frequency of the embedded permanent magnet synchronous motor, an online parameter identification method based on direct current signal injection is provided, which is used for calculating the flux linkage value and the equivalent loss resistance value of the fundamental frequency under the operating condition point. The method is suitable for the parameter identification application field of the embedded permanent magnet synchronous motor.

Description

Technical Field

[0001] This invention relates to an online parameter identification method for an embedded permanent magnet synchronous motor, belonging to the fields of electrical engineering, motor modeling, and motor control. Background Technology

[0002] Embedded permanent magnet synchronous motors, due to their advantages such as high torque density, wide speed range, fast dynamic response, and high constant power ratio, are widely used in many fields, including rail transportation, aerospace, household motors, and wind power generation. In these industrial applications, accurate motor parameters ensure efficient and stable motor operation. However, these parameters change with different operating conditions, leading to control errors. Motor parameter identification technology involves measuring the actual operating conditions of the motor and calculating and adjusting its parameters. Therefore, improving and refining motor parameter identification methods helps ensure motor control performance, increase operating efficiency, reduce failure rates, and facilitate motor debugging and maintenance.

[0003] In the field of motor parameter identification, current methods are mainly divided into offline and online parameter identification. Offline parameter identification methods primarily rely on finite element analysis and experimental estimation, but most are limited by the difficulty in obtaining material properties and cannot provide accurate parameters for the motor in real time. Online parameter identification methods, under specific operating conditions of voltage, current, and speed, detect relevant motor parameters in real time to monitor the motor's control performance and operating status. For embedded permanent magnet synchronous motors, since they inherently have only two steady-state voltage and current equations while there are many parameters to be estimated, signal injection methods are often used to increase the number of linearly independent equations, thereby establishing a full-rank reference model to achieve multi-parameter identification. Currently, commonly used signal injection methods are broadly classified into two categories: sinusoidal signal injection and DC signal injection. Existing sinusoidal signal injection methods use single-frequency signal injection, which cannot achieve synchronous identification when there are too many parameters to identify; furthermore, sinusoidal signal injection ignores the difference between the resistance at the injection frequency and the fundamental frequency resistance of the target solution, and its identification accuracy still needs further improvement. The DC signal injection method requires multiple steps to achieve multi-parameter identification, which has a deficiency in real-time performance. Summary of the Invention

[0004] The purpose of this invention is to propose an online parameter identification method for embedded permanent magnet synchronous motors based on hybrid signal injection, taking into account the difference between the fundamental frequency resistance and the high frequency resistance of the target solution. This method solves the underrank problem in the identification process of the intrinsic electrical model of embedded permanent magnet synchronous motors, improves the accuracy of parameter identification, and improves the multi-parameter synchronous identification problem that is difficult to achieve with existing online parameter identification methods, thereby increasing the parameter identification speed.

[0005] This invention, based on a mathematical model of an embedded permanent magnet synchronous motor considering the difference between fundamental frequency resistance and high-frequency resistance, proposes an online parameter identification method based on the hybrid injection of high-frequency sinusoidal current signals and DC current signals. First, sinusoidal current signals of different frequencies are simultaneously injected into the direct axis and quadrature axis to identify the incremental self-inductance and incremental mutual inductance at the operating point, overcoming the limitation of traditional single-frequency sinusoidal signal injection methods in achieving simultaneous multi-parameter identification. Simultaneously, a DC current signal is periodically injected into the direct axis to identify the fundamental frequency equivalent loss resistance and flux linkage values ​​at the target operating point. Only two steps are required to achieve complete and accurate identification of the electrical parameters of the embedded permanent magnet synchronous motor.

[0006] To solve the above problems, the technical solution adopted by the present invention is as follows:

[0007] First, a mathematical model of the embedded permanent magnet synchronous motor considering the difference between the base frequency resistance and the high-frequency resistance is established, including a steady-state incremental model of the base frequency and a small-signal dynamic model of the high frequency. Specifically:

[0008] (1) Steady-state incremental model of fundamental frequency:

[0009]

[0010]

[0011] In the above equation, i d0 and i q0 These are the direct-axis current base value and the quadrature-axis current base value, respectively, Δi d and Δi q These are the changes in direct-axis current and quadrature-axis current, respectively. When the direct-axis current and quadrature-axis current of the motor are respectively i d0 and i q0 At that time, the voltage at the direct shaft end and the voltage at the quadrature shaft end of the motor are defined as u. d0 and u q0 When the direct-axis current and quadrature-axis current of the motor are respectively i d0 +Δi d and i q0 +Δi q At that time, the voltage at the direct shaft end and the voltage at the quadrature shaft end of the motor are defined as u. d1 and u q1 ;Δu d and Δu q Defined respectively as the voltage change at the direct axis and the voltage change at the quadrature axis; R s ψ is the equivalent loss resistance value at the fundamental frequency; d0 and ψ q0 These represent the base values ​​of the direct-axis flux linkage and the quadrature-axis flux linkage of the motor, respectively; ω is the angular velocity of the motor; L dd and L qqThese are the direct-axis incremental self-inductance and quadrature-axis incremental self-inductance of the motor, respectively; L dq and L qd These are the direct-axis and quadrature-axis incremental mutual inductances of the motor, respectively.

[0012] (2) High-frequency small-signal dynamic model:

[0013]

[0014] In the above equation, i dh and i qh These are the injected direct-axis high-frequency small-current signal and the quadrature-axis high-frequency small-current signal, respectively; u dh and u qh These are direct-axis high-frequency low-voltage signals and quadrature-axis high-frequency low-voltage signals, respectively; R sh Equivalent loss resistor injected for high-frequency, low-current signals.

[0015] Subsequently, based on the established high-frequency small-signal dynamic model, an online parameter identification method based on dual-frequency sinusoidal current signal injection was used to calculate the incremental self-inductance and incremental mutual inductance of the permanent magnet synchronous motor. The specific steps are as follows:

[0016] (1) First, high-frequency sinusoidal current signals i of different frequencies are simultaneously injected into the direct axis and the quadrature axis. dh and i qh :

[0017]

[0018] Among them, I md and ω dh For high-frequency current signal i dh The amplitude and angular frequency, I mq and ω qh For high-frequency current signal i qh Amplitude and angular frequency;

[0019] (2) Based on the high-frequency small-signal dynamic model of the permanent magnet synchronous motor, we can obtain the result with ω dh Voltage and current equations at the same frequency:

[0020]

[0021] Among them, u ddh and u qdh These are the corresponding angular frequencies ω dh The direct-axis voltage signal and the quadrature-axis voltage signal.

[0022] (3) Change i dh Substituting the amplitude and angular frequency into step (2), we get:

[0023]

[0024] (4) Based on the analysis results of step (3), multiply the direct-axis voltage and quadrature-axis voltage by a frequency of ω respectively. dh The unit sinusoidal signal sin(ω) dh t), and through a low-pass filter, the direct-axis DC voltage U is obtained. ddh0 With cross-axis DC voltage U qdh0 :

[0025]

[0026] (5) Based on direct-axis DC voltage U ddh0 With cross-axis DC voltage U qdh0 The direct-axis incremental self-inductance value L can be calculated. dd Incremental mutual inductance value L of the perpendicular axis qd :

[0027]

[0028] (6) Based on the high-frequency small-signal dynamic model of the permanent magnet synchronous motor, we can obtain the result with ω qh Voltage and current equations at the same frequency:

[0029]

[0030] Among them, u dqh and u qqh These are the corresponding angular frequencies ω qh The direct-axis voltage signal and the quadrature-axis voltage signal.

[0031] (7) Change i qh Substituting the amplitude and angular frequency into step (6), we get:

[0032]

[0033] (8) Based on the analysis results of step (7), multiply the direct-axis voltage and quadrature-axis voltage by a frequency of ω respectively. qh The unit sinusoidal signal sin(ω) qh t), and through a low-pass filter, the direct-axis DC voltage U is obtained. dqh0 With cross-axis DC voltage U qqh0 :

[0034]

[0035] (9) Based on direct-axis DC voltage U dqh0 With cross-axis DC voltage U qqh0 The cross-axis incremental self-inductance value L can be calculated. qq Incremental mutual inductance value L of the orthogonal axis dq :

[0036]

[0037] Finally, based on the established fundamental frequency steady-state incremental model, and using the incremental inductance value obtained by the online parameter identification method based on dual-frequency sinusoidal current signal injection, the flux linkage value and fundamental frequency equivalent loss resistance value of the motor are calculated using the online parameter identification method based on DC current signal injection. The specific steps are as follows:

[0038] (1) Control the direct-axis current and quadrature-axis current of the motor to be i d0 and i q0 At this time, the DC voltage and DC current of the motor satisfy the equation:

[0039]

[0040] (2) Control the direct-axis current and quadrature-axis current of the motor as i d1 =i d0 +Δi d and i q1 =i q0 At this time, the DC voltage and DC current of the motor satisfy the equation:

[0041]

[0042] (3) Based on the two sets of voltage and current equations under DC steady-state conditions obtained in steps (1) and (2), and the quadrature-direct axis incremental mutual inductance value L calculated by the online parameter identification method based on dual-frequency sinusoidal current signal injection. qd It can calculate the equivalent loss resistance value R of the motor at its base frequency. s Direct-axis flux linkage base value ψ d0 and cross-axis flux linkage base value ψ q0 :

[0043]

[0044] The inventive principle of this invention is as follows:

[0045] First, this invention starts from the loss mechanism of embedded permanent magnet synchronous motors. Considering that the iron loss of the motor is positively correlated with frequency, it can be deduced that the equivalent loss resistance of the motor differs at different frequencies. That is, the resistance value observed under high-frequency sinusoidal signal injection cannot be used as the fundamental frequency resistance value for the objective solution. Therefore, this invention proposes that the fundamental frequency resistance value can be accurately identified by periodically injecting a DC current signal at the steady-state operating point.

[0046] However, if only DC current signal injection is used, the problem of multi-step signal injection is encountered, which limits the parameter identification speed. To address this, this invention proposes a hybrid signal injection method, in which a high-frequency sinusoidal signal is injected to identify the incremental inductance, while the DC signal injection is only responsible for identifying the fundamental frequency equivalent loss resistance and flux linkage. This reduces the number of parameters and signal injection steps required by the DC signal injection method, and also improves the parameter identification speed.

[0047] Furthermore, based on traditional high-frequency sinusoidal signal injection, this invention proposes a dual-frequency injection method to independently decompose multiple sets of parameter identification equations, simplifying the complex calculation process of solving multivariate equations and improving computer operating efficiency. Moreover, dual-frequency sinusoidal signal injection can introduce a total of eight linearly independent equations, twice that of the traditional single-frequency sinusoidal signal injection method, thus solving the problem of multi-parameter synchronous real-time identification that traditional methods cannot achieve.

[0048] The beneficial effects of this invention are as follows:

[0049] 1. The difference between the high-frequency resistance of the high-frequency small-signal dynamic model of the embedded permanent magnet synchronous motor and the fundamental frequency resistance of the target solution is considered. The fundamental frequency resistance is observed at the stable operating point by injecting DC current signal, which makes the identification result of the equivalent loss resistance value of the embedded permanent magnet synchronous motor more accurate.

[0050] 2. A dual-frequency current signal injection method is proposed in the high-frequency sinusoidal signal injection process, which introduces a total of eight linearly independent equations, twice that of the traditional single-frequency sinusoidal signal injection method. Therefore, it can solve the problem of multi-parameter synchronous real-time identification that the traditional method cannot achieve.

[0051] 3. This invention proposes a hybrid signal injection method, in which a high-frequency sinusoidal signal injection is responsible for identifying the incremental inductance, while the DC signal injection is only responsible for identifying the fundamental frequency equivalent loss resistance and flux linkage. This reduces the number of parameters and signal injection steps required for DC signal injection, thereby improving the parameter identification speed. Attached Figure Description

[0052] Figure 1 A schematic diagram of the mixed signal injection;

[0053] Figure 2 Equivalent circuit diagram of the fundamental frequency steady-state incremental model;

[0054] Figure 3 Equivalent circuit diagram of high-frequency small-signal dynamic model;

[0055] Figure 4 A flowchart of the method of the present invention. Detailed Implementation

[0056] The present invention will be further described below with reference to the accompanying drawings and specific embodiments.

[0057] Figure 1 A schematic diagram of the signals injected into the mixed signal illustrates the relationship between the injected signals, specifically:

[0058] (1) Direct-axis current signals include: high-frequency sinusoidal current signal i dh This is related to a periodic DC current signal. Specifically, the high-frequency sinusoidal current signal i... dh The amplitude and frequency remain constant under different direct-axis current base values; the DC current periodic signal, on the other hand, operates according to a fixed period, based on the direct-axis current base value i. d0 Add Δi in turn d And a bias of 0;

[0059] (2) The quadrature-axis current signal includes: high-frequency sinusoidal current signal i qh The sinusoidal current signal maintains a constant amplitude and frequency, and the quadrature-axis current base value also remains stable at i. q0 .

[0060] Figure 4 A flowchart illustrating an online parameter identification method for an embedded permanent magnet synchronous motor with hybrid signal injection according to the present invention is shown.

[0061] First, considering the difference between the fundamental frequency resistance and the high-frequency resistance, this invention proposes the following: a fundamental frequency steady-state incremental model (corresponding equivalent circuit diagram as shown in Figure 1). Figure 2 (As shown), high-frequency small-signal dynamic model (corresponding equivalent circuit diagram as shown) Figure 3 As shown). Specifically:

[0062] (1) Steady-state incremental model of fundamental frequency:

[0063]

[0064]

[0065] In the above equation, i d0 and i q0 These are the direct-axis current base value and the quadrature-axis current base value, respectively, Δi d and Δi q These are the changes in direct-axis current and quadrature-axis current, respectively. When the direct-axis current and quadrature-axis current of the motor are respectively i d0 and i q0 At that time, the voltage at the direct shaft end and the voltage at the quadrature shaft end of the motor are defined as u. d0 and u q0 When the direct-axis current and quadrature-axis current of the motor are respectively i d0 +Δi d and i q0 +Δiq At that time, the voltage at the direct shaft end and the voltage at the quadrature shaft end of the motor are defined as u. d1 and u q1 ;Δu d and Δu q Defined respectively as the voltage change at the direct axis and the voltage change at the quadrature axis; R s ψ is the equivalent loss resistance value at the fundamental frequency; d0 and ψ q0 These represent the base values ​​of the direct-axis flux linkage and the quadrature-axis flux linkage of the motor, respectively; ω is the angular velocity of the motor; L dd and L qq These are the direct-axis incremental self-inductance and quadrature-axis incremental self-inductance of the motor, respectively; L dq and L qd These are the direct-axis and quadrature-axis incremental mutual inductances of the motor, respectively.

[0066] (2) High-frequency small-signal dynamic model

[0067]

[0068] In the above equation, i dh and i qh These are the injected direct-axis high-frequency small-current signal and the quadrature-axis high-frequency small-current signal, respectively; u dh and u qh These are direct-axis high-frequency low-voltage signals and quadrature-axis high-frequency low-voltage signals, respectively; R sh The equivalent loss resistance injected for high-frequency, low-current signals. Here, "high-frequency" refers to the frequency of the injected sinusoidal current signal being 2 to 5 times the fundamental frequency of the motor; "low-frequency" refers to the amplitude of the injected sinusoidal current signal, approximately 0.1 to 5% of the fundamental current amplitude of the motor, with the specific value determined based on the actual application.

[0069] Next, based on the established high-frequency small-signal dynamic model, an online parameter identification method based on dual-frequency sinusoidal current signal injection is used to calculate the incremental self-inductance and incremental mutual inductance of the permanent magnet synchronous motor. The specific steps are as follows:

[0070] (1) First, high-frequency sinusoidal current signals i of different frequencies are simultaneously injected into the direct axis and the quadrature axis. dh and i qh :

[0071]

[0072] Among them, I md and ω dh For high-frequency current signal i dh The amplitude and angular frequency, I mq and ω qh For high-frequency current signal iqh Amplitude and angular frequency;

[0073] (2) Based on the high-frequency small-signal dynamic model of the permanent magnet synchronous motor, we can obtain the result with ω dh Voltage and current equations at the same frequency:

[0074]

[0075] Among them, u ddh and u qdh These are the corresponding angular frequencies ω dh The direct-axis voltage signal and the quadrature-axis voltage signal.

[0076] (3) Change i dh Substituting the amplitude and angular frequency into step (2), we get:

[0077]

[0078] (4) Based on the analysis results of step (3), multiply the direct-axis voltage and quadrature-axis voltage by a frequency of ω respectively. dh The unit sinusoidal signal sin(ω) dh t), and through a low-pass filter, the direct-axis DC voltage U is obtained. ddh0 With cross-axis DC voltage U qdh0 :

[0079]

[0080] (5) Based on direct-axis DC voltage U ddh0 With cross-axis DC voltage U qdh0 The direct-axis incremental self-inductance value L can be calculated. dd Incremental mutual inductance value L of the perpendicular axis qd :

[0081]

[0082] (6) Based on the high-frequency small-signal dynamic model of the permanent magnet synchronous motor, we can obtain the result with ω qh Voltage and current equations at the same frequency:

[0083]

[0084] Among them, u dqh and u qqh These are the corresponding angular frequencies ω qh The direct-axis voltage signal and the quadrature-axis voltage signal.

[0085] (7) Change i qh Substituting the amplitude and angular frequency into step (6), we get:

[0086]

[0087] (8) Based on the analysis results of step (7), multiply the direct-axis voltage and quadrature-axis voltage by a frequency of ω respectively. qh The unit sinusoidal signal sin(ω) qh t), and through a low-pass filter, the direct-axis DC voltage U is obtained. dqh0 With cross-axis DC voltage U qqh0 :

[0088]

[0089] (9) Based on direct-axis DC voltage U dqh0 With cross-axis DC voltage U qqh0 The cross-axis incremental self-inductance value L can be calculated. qq Incremental mutual inductance value L of the orthogonal axis dq :

[0090]

[0091] In steps (4) and (8), the selection rule for the cutoff frequency of the low-pass filter can be determined according to the actual situation. In practical applications, considering both data processing speed and filtering performance, a reasonable cutoff frequency should be selected, which is usually less than 1 / 5 to 1 / 2 of the frequency of the injected signal.

[0092] Finally, based on the established fundamental frequency steady-state incremental model, and using the incremental inductance value obtained by the online parameter identification method based on dual-frequency sinusoidal current signal injection, the flux linkage value and fundamental frequency equivalent loss resistance value of the motor are calculated using the online parameter identification method based on DC current signal injection. The specific steps are as follows:

[0093] (1) Control the direct-axis current and quadrature-axis current of the motor to be i d0 and i q0 At this time, the DC voltage and DC current of the motor satisfy the equation:

[0094]

[0095] (2) Control the direct-axis current and quadrature-axis current of the motor as i d1 =i d0 +Δi d and i q1 =i q0 At this time, the DC voltage and DC current of the motor satisfy the equation:

[0096]

[0097] (3) Based on the two sets of voltage and current equations under DC steady-state conditions obtained in steps (1) and (2), and the quadrature-direct axis incremental mutual inductance value L observed by the online parameter identification method based on dual-frequency sinusoidal current signal injection. qd It can calculate the equivalent loss resistance value R of the motor at its base frequency. s Direct-axis flux linkage base value ψ d0 and cross-axis flux linkage base value ψ q0 :

[0098]

Claims

1. A method for online parameter identification of an embedded permanent magnet synchronous motor with hybrid signal injection, characterized in that, Includes the following steps: Step 1: Establish a mathematical model of the embedded permanent magnet synchronous motor that considers the difference between the base frequency resistance and the high frequency resistance, including the steady-state incremental model of the base frequency and the small-signal dynamic model of the high frequency. Step 2: Based on the high-frequency small-signal dynamic model established in Step 1, the incremental self-inductance and incremental mutual inductance of the permanent magnet synchronous motor are calculated using the online parameter identification method based on dual-frequency sinusoidal current signal injection. Step 3: Based on the fundamental frequency steady-state incremental model established in Step 1, and using the incremental inductance value obtained in Step 2, the flux linkage value and fundamental frequency equivalent loss resistance value of the motor are calculated using the online parameter identification method based on DC current signal injection. The steady-state incremental model of the fundamental frequency and the small-signal dynamic model of the high frequency mentioned in step 1 are specifically as follows: (1) The steady-state incremental model of the fundamental frequency is: In the above equation, i d0 and i q0 These are the direct-axis current base value and the quadrature-axis current base value, respectively, Δi d and Δi q These are the changes in direct-axis current and quadrature-axis current, respectively; when the direct-axis current and quadrature-axis current of the motor are i d0 and i q0 At that time, the voltage at the direct shaft end and the voltage at the quadrature shaft end of the motor are defined as u. d0 and u q0 When the direct-axis current and quadrature-axis current of the motor are respectively i d0 +Δi d and i q0 +Δi q At that time, the voltage at the direct shaft end and the voltage at the quadrature shaft end of the motor are defined as u. d1 and u q1 ;Δu d and Δu q Defined respectively as the voltage change at the direct axis and the voltage change at the quadrature axis; R s ψ is the equivalent loss resistance value at the fundamental frequency; d0 and ψ q0 These represent the base values ​​of the direct-axis flux linkage and the quadrature-axis flux linkage of the motor, respectively; ω is the angular velocity of the motor; L dd and L qq These are the direct-axis incremental self-inductance and quadrature-axis incremental self-inductance of the motor, respectively; L dq and L qd These are the quadrature-direction incremental mutual inductance and the quadrature-direction incremental mutual inductance of the motor, respectively. (2) The high-frequency small-signal dynamic model is: In the above equation, i dh and i qh These are the injected direct-axis high-frequency small-current signal and the quadrature-axis high-frequency small-current signal, respectively; u dh and u qh These are direct-axis high-frequency low-voltage signals and quadrature-axis high-frequency low-voltage signals, respectively; R sh Equivalent loss resistance injected for high-frequency, low-current signals; In step 2, the direct-axis voltage and quadrature-axis voltage are multiplied by a frequency of ω. qh The unit sinusoidal signal sin(ω) qh t), and through a low-pass filter, the direct-axis DC voltage U is obtained. dqh0 With cross-axis DC voltage U qqh0 : Among them, I mq and ω qh For high-frequency current signal i qh Amplitude and angular frequency; Based on direct-axis DC voltage U dqh0 With cross-axis DC voltage U qqh0 The cross-axis incremental self-inductance value L was calculated. qq Incremental mutual inductance value L of the orthogonal axis dq : In step 3, based on the voltage and current equations under two sets of DC steady-state conditions and the incremental mutual inductance values ​​L on the direct and quadrature axes... qd The equivalent loss resistance value R of the motor at its fundamental frequency was calculated. s Direct-axis flux linkage base value ψ d0 and cross-axis flux linkage base value ψ q0 : In one set of voltage-current equations under DC steady-state conditions, the direct-axis current and quadrature-axis current of the control motor are i d1 =i d0 +Δi d and i q1 =i q0 The DC voltage and quadrature axis voltage of the motor are u q1 .

2. The online parameter identification method for an embedded permanent magnet synchronous motor with hybrid signal injection according to claim 1, characterized in that, The online parameter identification method based on dual-frequency sinusoidal current signal injection described in step 2 includes the following steps: (1) Simultaneously inject high-frequency sinusoidal current signals i of different frequencies into the direct axis and the quadrature axis respectively. dh and i qh : Among them, I md and ω dh For high-frequency current signal i dh Amplitude and angular frequency; (2) Based on the high-frequency small-signal dynamic model of the permanent magnet synchronous motor, the relationship with ω is obtained. dh Voltage and current equations at the same frequency: Among them, u ddh and u qdh These are the corresponding angular frequencies ω dh Direct-axis voltage signal and quadrature-axis voltage signal; (3) Change i dh Amplitude I md With angular frequency ω dh Substituting into step (2), we get: (4) Based on the analysis results of step (3), multiply the direct-axis voltage and quadrature-axis voltage by a frequency of ω respectively. dh The unit sinusoidal signal sin(ω) dh t), and through a low-pass filter, the direct-axis DC voltage U is obtained. ddh0 With cross-axis DC voltage U qdh0 : (5) Based on direct-axis DC voltage U ddh0 With cross-axis DC voltage U qdh0 The direct-axis incremental self-inductance value L was calculated. dd Incremental mutual inductance value L of the perpendicular axis qd : (6) Based on the high-frequency small-signal dynamic model of the permanent magnet synchronous motor, the relationship with ω is obtained. qh Voltage and current equations at the same frequency: Among them, u dqh and u qqh These are the corresponding angular frequencies ω qh Direct-axis voltage signal and quadrature-axis voltage signal; (7) Change i qh Amplitude I mq With angular frequency ω qh Substituting into step (6), we get: 。 3. The online parameter identification method for an embedded permanent magnet synchronous motor with hybrid signal injection according to claim 2, characterized in that, The online parameter identification method based on DC current signal injection described in step 3 constructs two sets of voltage and current equations under DC steady-state conditions. The specific steps are as follows: (1) Control the direct-axis current and quadrature-axis current of the motor as i d0 and i q0 At this time, the DC voltage and DC current of the motor satisfy the equation: (2) Control the direct-axis current and quadrature-axis current of the motor as i d1 =i d0 +Δi d and i q1 =i q0 At this time, the DC voltage and DC current of the motor satisfy the equation: 。

Images