A stator flux linkage observer and method of a permanent magnet assisted synchronous reluctance motor based on integral error estimation

Abstract

The application discloses a stator flux linkage observer and method of a permanent magnet auxiliary synchronous reluctance motor based on integral error estimation, and belongs to the technical field of motor control. The application is used for solving the problem that integral error in the existing stator flux linkage observer is difficult to be accurately estimated and purely eliminated without depending on accurate parameters. The application comprises the following steps: an integral module is used for performing integral operation on the stator flux linkage according to a voltage model of the motor, and a preliminary flux linkage estimation value containing integral error is output; an integral error observation module is arranged in parallel with the integral module, and is used for estimating an error component generated in the integral process in real time based on a current signal of the motor and a nominal inductance parameter through a time domain state observer; and a compensation module is connected with the integral module and the integral error observation module respectively, and is used for subtracting the error component estimated by the integral error observation module from the preliminary flux linkage estimation value output by the integral module, so as to output a stator flux linkage estimation value without direct current bias.

Description

Technical Field

[0001] This invention belongs to the field of motor control technology, specifically relating to a stator flux linkage observer and method for permanent magnet assisted synchronous reluctance motors, which is particularly suitable for high-precision estimation of flux linkage and rotor position in sensorless control systems. Background Technology

[0002] Permanent magnet assisted synchronous reluctance motors (PMaSynRM) combine the advantages of permanent magnet synchronous motors and synchronous reluctance motors, offering significant advantages such as high efficiency, high power density, wide speed range, no rotor copper loss, and low torque ripple. They show broad application prospects in new energy vehicles, industrial servo systems, and home appliances. Especially given the increasing scarcity of rare earth resources, permanent magnet assisted synchronous reluctance motors are considered one of the ideal alternatives to rare earth permanent magnet synchronous motors due to their lower usage of rare earth permanent magnet materials.

[0003] In high-performance motor drive systems, accurate rotor position information is fundamental for achieving field-oriented control and speed closed-loop control. Traditional methods typically use mechanical position sensors (such as photoelectric encoders and rotary transformers) to obtain rotor position, but these sensors have the following inherent drawbacks: (1) increased system cost and size; (2) decreased reliability under harsh conditions such as high temperature, high humidity, and vibration; (3) sensor signals are susceptible to electromagnetic interference; and (4) complex installation and maintenance. Therefore, sensorless control technology has become a research hotspot in the field of motor drives.

[0004] Accurate estimation of stator flux linkage is crucial for achieving sensorless control. Currently, there are two main models for stator flux linkage estimation: the voltage model and the current model. The voltage model estimates the flux linkage by integrating the stator voltage equation, and its mathematical form is:

[0005]

[0006] , These are the αβ-axis voltage matrix and αβ-axis current matrix in the stationary coordinate system, respectively. This is the stator resistance.

[0007] However, voltage models face a key technical challenge in practical applications: the accumulation of integral errors. Due to the DC bias in the integrator input signal (mainly caused by the measurement bias of voltage and current sensors) and inaccurate initial integration values, pure integrators will generate an integral error that accumulates over time, manifesting as a DC offset in the estimated flux linkage. This error leads to distortion of the flux linkage estimate, which in turn causes rotor position estimation errors, and in severe cases, even instability of the control system.

[0008] To eliminate integration errors, researchers have proposed several solutions, mainly including the following categories:

[0009] (1) Frequency domain filtering method

[0010] The most traditional approach is to use a high-pass filter instead of a pure integrator to suppress integration errors by filtering out the DC component. However, while suppressing the DC component, the high-pass filter also attenuates the amplitude and leads the phase of other frequency components in the flux linkage signal, causing distortion in the estimated amplitude and phase of the flux linkage and introducing additional angle estimation errors during dynamic processes. Although the low-pass filter with compensation method improves performance to some extent, it is essentially a frequency domain processing method and cannot completely avoid amplitude and phase distortion problems.

[0011] (2) Closed-loop observer method

[0012] To address the limitations of frequency domain methods, researchers have proposed various closed-loop observers, such as the Luneburger observer, sliding mode observer, and Kalman filter. These methods achieve flux estimation and error suppression by constructing an error feedback closed loop. However, traditional closed-loop observers typically rely on precise motor parameters (such as inductance and flux linkage) to construct the error feedback signal, making them highly sensitive to parameter errors. During actual motor operation, factors such as magnetic circuit saturation and temperature variations can cause significant changes in motor parameters, leading to a deterioration in observer performance. Even if offline measured motor parameter tables are available within a certain accuracy range, rotor position estimation errors during dynamic processes can still cause the table lookup values ​​to deviate from the actual operating conditions, resulting in observer parameter mismatch problems.

[0013] (3) High-frequency signal injection method

[0014] In the zero-speed and low-speed regions, the high-frequency signal injection method based on the salient pole effect of the motor is widely used for rotor position estimation. This method extracts rotor position information by injecting a high-frequency voltage signal into the stator winding and detecting the corresponding high-frequency current response. However, the high-frequency injection method has the following drawbacks: (1) the injected high-frequency signal introduces additional losses and torque ripple; (2) it requires a complex signal demodulation algorithm; and (3) it is only applicable to the zero-speed to low-speed region. In the medium- and high-speed regions, the signal-to-noise ratio decreases, requiring a switch to other methods. Therefore, in the medium- and high-speed regions, the flux linkage observer-based method is more suitable.

[0015] (4) Hybrid model method

[0016] In recent years, researchers have attempted to combine voltage and current models, leveraging their complementary properties to construct hybrid model flux observers. These methods typically employ a voltage model as the primary model in the medium-to-high speed range and a current model as the primary model in the low-speed range, achieving full-speed domain coverage through smooth switching. However, the design of the switching logic and the smooth transition between the two models remain technical challenges, and the current model also relies on accurate inductance parameters.

[0017] In summary, existing technologies for addressing integral error either introduce amplitude and phase distortion (frequency domain methods), rely on precise parameters (closed-loop observers), or incur additional losses (high-frequency injection methods). Therefore, there is an urgent need for a stator flux linkage observation method that does not rely on precise parameters, does not distort signals, and has a fast dynamic response, in order to achieve high-performance sensorless control of permanent magnet assisted synchronous reluctance motors. Summary of the Invention

[0018] To address the problem that integral errors in existing stator flux linkage observers are difficult to accurately estimate and completely eliminate without relying on precise parameters, this invention provides a stator flux linkage observer and method for permanent magnet assisted synchronous reluctance motors based on integral error estimation.

[0019] In a first aspect, the present invention provides a stator flux linkage observer for a permanent magnet assisted synchronous reluctance motor based on integral error estimation, comprising:

[0020] The integration module is used to perform integral calculations on the stator flux linkage based on the motor's voltage model and output a preliminary flux linkage estimate including integration error.

[0021] An integral error observation module, set in parallel with the integral module, is used to estimate the error components generated during the integration process in real time through a time-domain state observer based on the motor's current signal and nominal inductance parameters.

[0022] The compensation module is connected to the integration module and the integration error observation module respectively, and is used to subtract the error component estimated by the integration error observation module from the preliminary flux estimate output by the integration module, and output the stator flux estimate without DC bias.

[0023] Preferably, the integration module includes:

[0024] The flux linkage rate calculation unit is used to calculate the αβ axis voltage matrix in the stationary coordinate system. α-β axis current matrix and stator resistance Calculate the rate of change of magnetic flux :

[0025] ;

[0026] The integration unit, connected to the flux linkage rate calculation unit, is used to perform integration calculations on the flux linkage rate and output a preliminary flux linkage estimate in the stationary coordinate system. :

[0027]

[0028] Among them, the preliminary flux linkage estimate Includes integration error caused by DC bias or inaccurate initial values. .

[0029] Preferably, the integral error observation module includes:

[0030] The input processing unit is used to receive the αβ axis current matrix in the stationary coordinate system. and nominal inductance parameters ;

[0031] The state observer core, connected to the input processing unit, is used to construct a time-domain linear observer based on a state-space model.

[0032] The Kalman filter, integrated into the core of the state observer, is used to achieve optimal estimation of the state vector.

[0033] Preferably, the state-space model is represented as:

[0034]

[0035] In the formula, For state vectors, , For the αβ axis flux linkage error, It is a velocity vector; State vector The derivative;

[0036] For the output vector, ;

[0037] For the system matrix, , Represents a 2x2 zero matrix. Represents a 2x2 identity matrix. The rotor's mechanical angular velocity, for Rotation matrix, ;

[0038] For the output matrix, .

[0039] Preferably, the Kalman filter outputs three estimated parameters in each sampling period:

[0040] integral error estimate The output is sent to the compensation module;

[0041] αβ axis flux linkage error estimate This is caused by inductance error and permanent magnet flux linkage;

[0042] Velocity vector estimate ;

[0043] The three estimated parameters , , Together they form the state vector The estimated value This information is fed back to the input of the Kalman filter for state updates at the next time step.

[0044] Preferably, the compensation module includes:

[0045] The subtractor has its first input connected to the output of the integrator module, and receives the preliminary flux linkage estimate. Its second input terminal is connected to the output terminal of the integration error observation module to receive the integration error estimate. ;

[0046] The compensation arithmetic unit is used to perform subtraction operations. Output stator flux linkage estimate without DC bias .

[0047] In a second aspect, the present invention provides a method for observing the stator flux linkage of a permanent magnet assisted synchronous reluctance motor based on integral error estimation, the method comprising the following steps:

[0048] Acquire the αβ axis voltage matrix in the stationary coordinate system of the motor α-β axis current matrix ;

[0049] The flux change rate was measured using an integrator module. Integrating, we obtain a preliminary flux linkage estimate that includes integration error. ;

[0050] Through the integration error observation module, based on the αβ axis current matrix and nominal inductance parameters Construct a state-space model;

[0051] A Kalman filter is used to perform optimal estimation of the state vector, and three estimation parameters are output in each sampling period: the integral error estimate. αβ axis flux linkage error estimate Velocity vector estimate ;

[0052] The three estimated parameters , , Feedback is sent to the input of the Kalman filter to form the state vector estimate. This is used for updating the state in the next moment;

[0053] The integral error estimate is adjusted using the compensation module. From preliminary flux linkage estimates Subtract from the original value to reconstruct the stator flux estimate without DC bias. ;

[0054] Based on the stator flux linkage estimate Calculate the rotor position estimate ;

[0055] The rotor position estimate Input the phase-locked loop to obtain the rotor speed estimate. ;

[0056] The rotor speed estimate It serves as speed feedback and participates in the closed-loop speed control of the motor.

[0057] The beneficial effects of this invention are:

[0058] (1) Cleaner signal with error elimination (no signal distortion). Compared with the traditional frequency domain high-pass filter method, this time domain observer only estimates and subtracts the integration error (such as DC bias) without affecting other frequency components in the flux signal. This avoids the amplitude and phase distortions common in traditional methods and ensures the accuracy of flux estimation.

[0059] (2) Faster dynamic response. Because the error is processed directly in the time domain, rather than through a filter with hysteresis, this method can eliminate integral errors more quickly. This allows the observer to maintain high estimation performance even during changes in motor load or transient processes.

[0060] (3) Insensitive to parameters (strong robustness). It is specifically pointed out at the end of the paper that the state observer used in this method does not require precise motor parameter information. This is different from other observer methods that rely on precise mathematical models, making it more robust in practical applications and less prone to performance degradation due to parameter changes or inaccurate measurements.

[0061] (4) Applicable to medium- and high-speed sensorless control. This method belongs to the model scheme based on flux observer and is specifically designed for the medium- and high-speed region. Compared with the high-frequency signal injection method, it avoids the additional losses caused by injecting high-frequency signals and improves system efficiency.

[0062] (5) The logic is clear and the implementation is straightforward. Its core idea is very intuitive—based on the traditional integration results, an observer is used to calculate the error and subtract it. This structure retains the framework of the traditional voltage model integration method and cleverly solves its inherent integration drift problem. Attached Figure Description

[0063] Figure 1 This is the effective magnetic flux vector diagram described in this invention;

[0064] Figure 2 A diagram showing the components of the integral magnetic flux linkage described in this invention;

[0065] Figure 3 This is a structural diagram of the stator flux linkage observer described in this invention;

[0066] Figure 4 This is a schematic diagram of the overall scheme of a sensorless control system for a permanent magnet assisted synchronous reluctance motor using the stator flux linkage observer described in this invention. Detailed Implementation

[0067] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0068] It should be noted that, unless otherwise specified, the embodiments and features described in the present invention can be combined with each other.

[0069] The present invention will be further described below with reference to the accompanying drawings and specific embodiments, but this is not intended to limit the scope of the invention.

[0070] In traditional approaches, high-frequency signal injection methods based on the motor salient pole model are typically only used for rotor position and speed estimation in the zero-speed to low-speed range. As speed increases, the signal-to-noise ratio of the flux linkage observer improves; therefore, using a flux linkage observer-based method for sensorless control in the medium-to-high-speed range is more suitable to avoid the additional losses caused by high-frequency injection. The flux linkage observer is designed based on voltage and current models of the flux linkage. The voltage model estimates the stator flux linkage by integrating the voltage equation, while the current model estimates the stator flux linkage by multiplying the inductance and current.

[0071] The mathematical forms of the voltage and current models of magnetic flux linkage are as follows:

[0072] (1)

[0073] in, Let be the αβ axis flux linkage matrix. , These are the α-axis flux linkage and the β-axis flux linkage, respectively. For the αβ axis voltage matrix, , These are the α-axis voltage and the β-axis voltage, respectively. For stator resistance, For the αβ axis current matrix, , These are the α-axis current and the β-axis current, respectively. For the αβ axis inductance matrix, , These are the α-axis inductance and the β-axis inductance, respectively. It is a permanent magnet flux linkage. The rotor position (characterized by rotor electrical angles). For the dq axis inductance matrix, , These are the d-axis inductance and the d-axis inductance, respectively. The rate of change of magnetic flux ( (first derivative) This is the coordinate transformation matrix.

[0074] Depend on Figure 1 As shown in the effective flux chain vector diagram, the effective flux chain is... The components in the αβ axis coordinate system are:

[0075] (2)

[0076] in, For effective magnetic flux Components along the α-axis, For effective magnetic flux The component along the β axis.

[0077] Effective flux linkage can also be represented as:

[0078] (3)

[0079] in, For α-axis flux linkage, For β-axis flux linkage, For the α-axis current, For β-axis current, For d-axis inductance, It is the q-axis inductance.

[0080] The estimated rotor electrical angle can be expressed as:

[0081] (4)

[0082] in, For the estimated rotor electrical angle, This is an estimate of the effective flux linkage α-axis component. This is an estimate of the effective flux linkage β-axis component. Figure 1 middle This is the stator current value.

[0083] The stator flux linkage estimated using a pure integral controller based on the voltage model is:

[0084] (5)

[0085] in, This is the DC integral offset.

[0086] In flux linkage estimation, stator resistance is usually assumed. Keep constant. Error term. The main causes of rotor position estimation errors stemming from DC bias in voltage or current signals and unknown initial integration conditions are distortions in the estimated flux linkage, leading to errors in rotor position estimation and deteriorating system control performance. To eliminate... Common methods include high-pass filters and closed-loop observers. However, while high-pass filters suppress DC components, they also cause amplitude and phase distortions in the estimated flux linkage, introducing angle estimation errors. Closed-loop observers typically rely on accurate inductance or flux linkage parameters to construct error feedback signals, making them highly sensitive to parameter errors. Even if offline measurement-based motor parameter tables are available within a certain accuracy range, rotor position estimation errors during dynamic processes can still cause table lookup values ​​to deviate from actual operating conditions, leading to observer parameter mismatch problems.

[0087] To address the problems inherent in traditional methods, this invention introduces a novel stator flux linkage observer that enhances dynamic performance without relying on conventional methods. The proposed estimator is based on the traditional method of integrating the stator flux linkage differential equation in the αβ reference frame, but its innovation lies in the method of eliminating integration errors. The core idea is to utilize a time-domain state observer to estimate the integration error and subtract it from the integration result. Compared to traditional frequency-domain methods that typically use high-pass filters, this time-domain method can eliminate integration errors more quickly. Notably, the state observer used in this method does not require precise parameter information, which distinguishes it from other state observer-based methods.

[0088] To avoid dependence on precise inductance parameters when constructing the feedback term, the flux linkage deviation caused by inductance error is further observed as an unknown state. The actual flux linkage can be expressed as the sum of the flux linkage generated by the nominal inductance and the flux linkage caused by the inductance error, and its expression is:

[0089] (6)

[0090] in, For the nominal inductance matrix, Here is the inductance error matrix. This refers to the αβ axis flux linkage error caused by inductance error and permanent magnet flux linkage.

[0091] The integral flux linkage can be reformulated as:

[0092] (7)

[0093] Under steady-state operating conditions of the motor, the current vector rotates uniformly in space, and its angular velocity is related to the rotor's mechanical angular velocity. Consistent. Correspondingly, the flux linkage vector... and its components and Also at the same rotor mechanical angular velocity Synchronous rotation. When a DC bias exists in the voltage or current signal, the error term... Along the velocity vector The direction of movement is then determined. At this point, the trajectory of the flux linkage vector obtained by integration appears as a path with... With the center, and A circle rotating at an angular velocity, such as Figure 2 As shown.

[0094] Example 1: This invention provides a stator flux linkage observer for a permanent magnet assisted synchronous reluctance motor based on integral error estimation. See [link to relevant documentation]. Figure 3 As shown, the observer includes: an integration module, an integration error observation module, and a compensation module.

[0095] I. Points Module

[0096] Integral module and Figure 3 The upper part of the path corresponds to the stator flux linkage, which is used to perform integral calculations on the stator flux linkage based on the motor's voltage model, and outputs a preliminary flux linkage estimate including integration error. .

[0097] Specifically, the integration module includes:

[0098] (1) Calculation unit for flux change rate

[0099] The flux linkage rate calculation unit receives the αβ axis voltage matrix in the stationary coordinate system. α-β axis current matrix and stator resistance The flux linkage rate is calculated according to equation (8):

[0100] (8)

[0101] In the formula, the term on the right side This is the back electromotive force, which, according to Faraday's law of electromagnetic induction, is equal to the time derivative of the magnetic flux linkage.

[0102] (2) Integral operation unit

[0103] The integration unit is connected to the flux linkage rate calculation unit and is used to perform integration on the flux linkage rate calculated by equation (8) and output the preliminary flux linkage estimate in the stationary coordinate system. :

[0104] (9)

[0105] exist Figure 3 In the process, the integral operation unit adopts a transfer function as follows: The integrator implementation.

[0106] Because voltage and current sensors in actual systems have measurement biases, and the initial integral value is difficult to determine accurately, the preliminary flux linkage estimate is... This will include integration errors caused by DC bias or inaccurate initial values. For details, please refer to equation (5). Integral error Under steady-state conditions, this manifests as a DC bias, which leads to distortion of the flux linkage estimate and consequently causes errors in rotor position estimation.

[0107] II. Integral Error Observation Module

[0108] The integration error observation module and the integration module are configured in parallel, such as... Figure 3 The lower half is shown, which is used to estimate the error components generated during the integration process in real time by using a time-domain state observer based on the motor's current signal and nominal inductance parameters.

[0109] The integral error observation module includes: an input processing unit, a state observer core, and a Kalman filter.

[0110] (1) Input processing unit

[0111] The input processing unit is used to receive the αβ axis current matrix in the stationary coordinate system. and nominal inductance parameters Among them, the nominal inductance parameters This is the nominal inductance value of the motor; a precise value is not required. It can be obtained through offline measurement or parameters provided by the manufacturer.

[0112] (2) State observer core

[0113] The state observer core is connected to the input processing unit and is used to construct a time-domain linear observer based on a state-space model.

[0114] According to the mathematical model of a permanent magnet assisted synchronous reluctance motor, under steady-state operating conditions, the current vector rotates uniformly in space, and its angular velocity is related to the rotor's mechanical angular velocity. Consistent. Correspondingly, the flux linkage vector... and its components and Also at the same rotor mechanical angular velocity Synchronous rotation. Based on this characteristic, a state-space model as shown in equation (10) is constructed:

[0115] (10)

[0116] In the formula, For state vectors, , For the αβ axis flux linkage error, It is a velocity vector; State vector The derivative;

[0117] For the output vector, (11)

[0118] For the system matrix, (12)

[0119] In equation (12):

[0120] Represents a 2x2 zero matrix. Represents a 2x2 identity matrix. The rotor's mechanical angular velocity, for Rotation matrix, , used to describe the rotational motion of a vector;

[0121] For the output matrix, , is a constant matrix.

[0122] The Kalman filter outputs three estimated parameters in each sampling period:

[0123] integral error estimate The output is sent to the compensation module;

[0124] αβ axis flux linkage error estimate This is caused by inductance error and permanent magnet flux linkage;

[0125] Velocity vector estimate ;

[0126] The three estimated parameters , , Together they form the state vector The estimated value This information is fed back to the input of the Kalman filter for state updates at the next time step.

[0127] III. Compensation Module

[0128] The compensation module is connected to both the integration module and the integration error observation module, and is used to process the integration error estimate obtained by the integration error observation module. The preliminary flux linkage estimate output from the integration module Subtract from the output to get the estimated stator flux linkage without DC bias. .

[0129] like Figure 3 As shown on the right, the compensation module includes:

[0130] (1) Subtractor

[0131] The first input of the subtractor is connected to the output of the integrator module to receive the preliminary flux linkage estimate. Its second input terminal is connected to the output terminal of the integration error observation module to receive the integration error estimate. .

[0132] exist Figure 3 In the diagram, the subtractor is represented by a circle (Σ) with a minus sign, which intuitively reflects its signal difference function.

[0133] (2) Compensation calculation unit

[0134] The compensation operation unit is connected to the subtractor and is used to perform the subtraction operation shown in equation (13):

[0135] (13)

[0136] The compensation operation unit can be an arithmetic logic unit in a digital signal processor, which implements subtraction operations through software code; or it can be a subtractor circuit in a hardware circuit.

[0137] The output of the compensation module is an estimated stator flux linkage without DC bias. This estimated value is the final estimated stator flux linkage, which can be directly used for subsequent rotor position calculations and speed estimations.

[0138] Explanation of the compensation principle:

[0139] Due to the initial flux linkage estimate output by the integrator module This includes integration errors caused by inaccurate DC bias or initial values, and the output of the integration error observation module... The integral error is obtained by real-time estimation of the Kalman filter. Subtracting the two will give the true flux linkage value.

[0140] Comparison with traditional methods:

[0141] and Figure 3 Unlike the traditional method of directly high-pass filtering the integration result in the upper part, the compensation module of this invention eliminates integration errors by subtraction rather than filtering. This method has the following advantages:

[0142] 1. Selective elimination: Only the error component is subtracted. This does not affect other frequency components in the flux linkage signal;

[0143] 2. No amplitude and phase distortion: Avoids amplitude attenuation and phase lead introduced by high-pass filters;

[0144] 3. Good real-time performance: Subtraction operation has no lag and fast dynamic response;

[0145] 4. Simple to implement: Only one subtraction operation is required, and the amount of calculation is small.

[0146] Example 2: This invention provides a method for observing the stator flux linkage of a permanent magnet assisted synchronous reluctance motor based on integral error estimation. The method includes the following steps:

[0147] Step 1: Signal Acquisition

[0148] Acquire the αβ axis voltage matrix in the stationary coordinate system of the motor α-β axis current matrix Voltage and current signals can be obtained in real time by voltage and current sensors and transformed into a stationary coordinate system through Clarke transformation.

[0149] Step 2: Preliminary flux linkage estimation

[0150] The flux change rate was measured using an integrator module. Integrating, we obtain a preliminary flux linkage estimate that includes integration error. :

[0151] (14)

[0152] Step 3: State-space model construction

[0153] Through the integration error observation module, based on the αβ axis current matrix and nominal inductance parameters Construct state-space models as shown in equations (10) to (12);

[0154] Step 4: Kalman filter estimation

[0155] A Kalman filter is used to perform optimal estimation of the state vector, and three estimation parameters are output in each sampling period: the integral error estimate. αβ axis flux linkage error estimate Velocity vector estimate ;

[0156] Step 5: Status Feedback Update

[0157] The three estimated parameters described in step 4 , , Feedback is sent to the input of the Kalman filter to form the state vector estimate. This is used for updating the state in the next moment;

[0158] Step 6: Integral Error Compensation

[0159] The integral error estimate is adjusted using the compensation module. From preliminary flux linkage estimates Subtract from the original value to reconstruct the stator flux estimate without DC bias. See equation (13) for details.

[0160] Step 7: Rotor position calculation

[0161] Based on the stator flux estimate obtained in step 6 Calculate the rotor position estimate ;

[0162] According to the formula Calculate the estimated value of the rotor electrical angle;

[0163] Step 8: Rotational Speed ​​Estimation

[0164] The rotor position estimate Input the phase-locked loop to obtain the rotor speed estimate. The phase-locked loop (PLL) achieves angle tracking and speed extraction through closed-loop regulation.

[0165] Step 9: Speed ​​Closed-Loop Control

[0166] The rotor speed estimate As speed feedback, with a given rotational speed The input to the speed controller is compared, and the output torque command is used to participate in the closed-loop speed control of the motor.

[0167] Example 3: Sensorless Control System

[0168] This embodiment provides a sensorless control system for a permanent magnet assisted synchronous reluctance motor, such as... Figure 4 As shown. The system includes the stator flux linkage observer described in Embodiment 1, as well as a position calculation module, a phase-locked loop, a speed controller, a maximum torque-to-current ratio control module, a current controller, a coordinate transformation module, and a PWM inverter.

[0169] (1) Stator flux linkage observer

[0170] The stator flux linkage observer, as described in Example 1, is used to measure the αβ-axis voltage matrix of the stator. α-β axis current matrix Output stator flux linkage estimate without DC bias .

[0171] (2) Location Calculation Module

[0172] The position calculation module is embedded in the stator flux linkage observer and is used to calculate the position based on the stator flux linkage estimate. Calculate the rotor position estimate .

[0173] (3) Phase-locked loop

[0174] The phase-locked loop is connected to the position calculation module embedded in the stator flux linkage observer, and is used to calculate the rotor position based on the estimated value. Estimated rotor speed The phase-locked loop (PLL) contains a phase detector, a loop filter, and a voltage-controlled oscillator (VCO) to track the rotor position and extract its rotational speed.

[0175] (4) Speed ​​controller

[0176] The speed controller is connected to a phase-locked loop (PLL) to control the speed according to the speed command. And rotor speed estimate The error is adjusted by regulating the output torque command via PI control.

[0177] (5) Maximum torque-to-current ratio control module

[0178] The Maximum Torque-to-Current Ratio (MTPA) control module is connected to the speed controller and is used to optimize the allocation of dq-axis current commands according to the MTPA trajectory based on the torque command, the current speed, and motor parameters. , This minimizes copper losses when the motor outputs the same torque.

[0179] (6) Current controller

[0180] The current controller is used to control the dq axis current command. , and feedback dq axis current , The error is controlled by adjusting the output dq axis voltage command via PI control. , .

[0181] (7) Coordinate Transformation Module

[0182] The coordinate transformation module includes Park transformation and inverse Park transformation:

[0183] Inverse Park transformation: converts the dq-axis voltage command , Transform into αβ axis voltage command , Using rotor position estimates ;

[0184] Park transformation: This involves converting the collected three-phase current of the motor into a single data source. , , Transformed into αβ axis current , Then the αβ axis current , Transformed into dq-axis feedback current , It is used for current closed-loop control.

[0185] (8) PWM inverter

[0186] PWM inverters are used to control the voltage of the α and β axes. , The system generates a three-phase pulse width modulation signal through space vector modulation (SVPWM) to drive a three-phase inverter to supply power to the motor.

[0187] Through the coordinated operation of the above modules, the system achieves high-performance sensorless control of permanent magnet assisted synchronous reluctance motors, with advantages such as strong parameter robustness, fast dynamic response, and no high-frequency injection loss.

[0188] The improved flux linkage observer proposed in this invention directly estimates and compensates for integration errors through a time-domain state observer, solving the flux linkage distortion problem caused by traditional frequency-domain methods. Its core advantage lies in its independence from precise motor parameters and excellent dynamic performance. This method can completely eliminate DC bias, achieve minimal rotor position estimation errors, and maintain stability under load variations. This achievement provides a more robust solution for sensorless control, especially in medium- and high-speed applications.

[0189] While the invention has been described herein with reference to specific embodiments, it should be understood that these embodiments are merely examples of the principles and applications of the invention. Therefore, it should be understood that many modifications can be made to the exemplary embodiments, and other arrangements can be designed without departing from the spirit and scope of the invention as defined by the appended claims. It should be understood that different dependent claims and features described herein can be combined in ways different from those described in the original claims. It is also understood that features described in conjunction with individual embodiments can be used in other described embodiments.

Claims

1. A stator flux linkage observer for a permanent magnet assisted synchronous reluctance motor based on integral error estimation, characterized in that, include: The integration module is used to perform integral calculations on the stator flux linkage based on the motor's voltage model and output a preliminary flux linkage estimate including integration error. An integral error observation module, set in parallel with the integral module, is used to estimate the error components generated during the integration process in real time through a time-domain state observer based on the motor's current signal and nominal inductance parameters. The compensation module is connected to the integration module and the integration error observation module respectively, and is used to subtract the error component estimated by the integration error observation module from the preliminary flux estimate output by the integration module, and output the stator flux estimate without DC bias.

2. The stator flux linkage observer for a permanent magnet assisted synchronous reluctance motor based on integral error estimation according to claim 1, characterized in that, The integration module includes: The flux linkage rate calculation unit is used to calculate the αβ axis voltage matrix in the stationary coordinate system. α-β axis current matrix and stator resistance Calculate the rate of change of magnetic flux : ； The integration unit, connected to the flux linkage rate calculation unit, is used to perform integration calculations on the flux linkage rate and output a preliminary flux linkage estimate in the stationary coordinate system. : Among them, the preliminary flux linkage estimate Includes integration error caused by DC bias or inaccurate initial values. .

3. The stator flux linkage observer for a permanent magnet assisted synchronous reluctance motor based on integral error estimation according to claim 2, characterized in that, The integral error observation module includes: The input processing unit is used to receive the αβ axis current matrix in the stationary coordinate system. and nominal inductance parameters ; The state observer core, connected to the input processing unit, is used to construct a time-domain linear observer based on a state-space model. The Kalman filter, integrated into the core of the state observer, is used to achieve optimal estimation of the state vector.

4. The stator flux linkage observer for a permanent magnet assisted synchronous reluctance motor based on integral error estimation according to claim 3, characterized in that, The state-space model is represented as follows: In the formula, For state vectors, , For the αβ axis flux linkage error, It is a velocity vector; State vector The derivative; For the output vector, ; For the system matrix, , Represents a 2x2 zero matrix. Represents a 2x2 identity matrix. The rotor's mechanical angular velocity, for Rotation matrix, ; For the output matrix, .

5. The stator flux linkage observer for a permanent magnet assisted synchronous reluctance motor based on integral error estimation according to claim 4, characterized in that, The Kalman filter outputs three estimated parameters in each sampling period: integral error estimate The output is sent to the compensation module; αβ axis flux linkage error estimate This is caused by inductance error and permanent magnet flux linkage; Velocity vector estimate ; The three estimated parameters , , Together they form the state vector The estimated value This information is fed back to the input of the Kalman filter for state updates at the next time step.

6. A stator flux linkage observer for a permanent magnet assisted synchronous reluctance motor based on integral error estimation according to claim 5, characterized in that, The compensation module includes: The subtractor has its first input connected to the output of the integrator module, and receives the preliminary flux linkage estimate. Its second input terminal is connected to the output terminal of the integration error observation module to receive the integration error estimate. ; The compensation arithmetic unit is used to perform subtraction operations. Output stator flux linkage estimate without DC bias .

7. A method for observing the stator flux linkage of a permanent magnet assisted synchronous reluctance motor based on integral error estimation, implemented using the stator flux linkage observer described in any one of claims 1-6, characterized in that, The method includes the following steps: Acquire the αβ axis voltage matrix in the stationary coordinate system of the motor α-β axis current matrix ; The flux change rate was measured using an integrator module. Integrating, we obtain a preliminary flux linkage estimate that includes integration error. ; Through the integration error observation module, based on the αβ axis current matrix and nominal inductance parameters Construct a state-space model; A Kalman filter is used to perform optimal estimation of the state vector, and three estimation parameters are output in each sampling period: the integral error estimate. αβ axis flux linkage error estimate Velocity vector estimate ; The three estimated parameters , , Feedback is sent to the input of the Kalman filter to form the state vector estimate. This is used for updating the state in the next moment; The integral error estimate is adjusted using the compensation module. From preliminary flux linkage estimates Subtract from the original value to reconstruct the stator flux estimate without DC bias. ; Based on the stator flux linkage estimate Calculate the rotor position estimate ; The rotor position estimate Input the phase-locked loop to obtain the rotor speed estimate. ; The rotor speed estimate It serves as speed feedback and participates in the closed-loop speed control of the motor.

Images