An improved flux linkage observer position sensorless observation method and system

Abstract

The application discloses an improved flux linkage observer position sensor observation method and system, comprising the following steps: constructing a motor model, and estimating a flux linkage vector; performing amplitude normalization processing on the flux linkage vector, and constructing a quadrature error signal; inputting a phase-locked loop, and obtaining a motor angle and a motor estimated rotating speed. The application has high adaptability, is small in system dependence degree, and is convenient to transplant in different systems.

Description

Technical Field

[0001] This invention relates to the field of motor control, and in particular to an improved magnetic flux observer without position sensor observation method and system. Background Technology

[0002] With the continuous development of modern permanent magnet synchronous motor drive technology, service life and cost have become key concerns for permanent magnet synchronous motor drive systems. Traditional permanent magnet synchronous motor drive systems rely heavily on bulky, expensive, and short-lived electrolytic capacitors in the rectification stage to stabilize the DC bus voltage, which has become a key bottleneck restricting further improvements in power density, service life, and cost of the drive system.

[0003] To overcome this bottleneck, electrolytic-capacitor-free PMSM drive systems (ECFPDS) that replace electrolytic capacitors with thin-film capacitors have been proposed in recent years. Compared with traditional structures, ECFPDS exhibits significant advantages in terms of size, lifespan, and cost, making it suitable for applications with stringent power density and cost requirements. However, removing the electrolytic capacitors significantly reduces the system's energy buffering capacity, causes periodic fluctuations in the DC bus voltage at integer multiples of the grid frequency, and introduces harmonic components at integer multiples of the grid frequency into the back electromotive force and flux linkage. This leads to a decrease in the accuracy of sensorless algorithm observations, periodic deviations in rotor position and angle estimations, and severely affects the system's robustness and dynamic performance.

[0004] Existing technologies, when addressing the aforementioned problems, primarily filter or compensate for the input signal of the observer. These methods have limited effectiveness and cannot fundamentally suppress periodic disturbances caused by DC bus voltage fluctuations. Furthermore, filtering easily introduces delays, further reducing the system's dynamic response speed and accuracy, failing to meet the requirements of high-performance drive systems. In addition, existing technologies have not directly incorporated resonant control concepts into the phase-locked loop structure, resulting in insufficient suppression of periodic disturbances. The system is susceptible to harmonic components, leading to accumulated observation errors and ultimately affecting the efficient and stable operation of the motor. Therefore, an innovative technical method is urgently needed to effectively improve the system's robustness and dynamic performance, solving the problem of periodic disturbances in electrolytic capacitor-free drive systems. Summary of the Invention

[0005] In order to overcome the above-mentioned shortcomings and deficiencies of the prior art, the purpose of this invention is to provide an improved magnetic flux observer without position sensor and an observation method and system, which is applicable to permanent magnet synchronous motor without electrolytic capacitor drive system.

[0006] The objective of this invention is achieved through the following technical solution:

[0007] An improved magnetic flux observer without a position sensor includes the following steps:

[0008] Construct a motor model and estimate the flux linkage vector;

[0009] The flux linkage vector is normalized to construct an orthogonal error signal.

[0010] Input the phase-locked loop to obtain the motor angle and the estimated motor speed.

[0011] Furthermore, the specific steps for constructing the motor model are as follows:

[0012] The stator current in the two-phase stationary coordinate system is obtained by performing Clark transform on the three-phase current of the motor. ;

[0013] Perform an inverse Park transformation on the output of the current controller to calculate the stator voltage in the two-phase stationary coordinate system. ;

[0014] In a stationary coordinate system, the stator voltage, stator current, and flux linkage vector satisfy the following:

[0015]

[0016]

[0017] in, For stator flux linkage, Let Ls be the stator flux linkage differential, and Ls and Rs be the inductance and resistance of the motor, respectively.

[0018] Furthermore, the estimated flux linkage vector is specifically estimated by observing the flux linkage vector using an improved nonlinear observation law.

[0019] Furthermore, the improved nonlinear observation law is specifically as follows:

[0020]

[0021] in, , Gradient gain; , The phase correction symbol has the following discrete form: , For rotor flux, It is a permanent magnet flux linkage.

[0022] Furthermore, the phase-locked loop is specifically a PIR controller with a resonant element.

[0023] Furthermore, the structure of the PIR controller with resonant loop is as follows:

[0024]

[0025] in, If it is a single-phase ECFPDS, then If it is a three-phase ECFPDS, then ; The damping coefficient; For resonant gain, , These are the proportional coefficient and the integral coefficient.

[0026] Furthermore, by inputting the phase-locked loop (PLL), the motor angle and the estimated rotor speed are obtained, specifically:

[0027] The phase-locked loop outputs a mechanical angular velocity, which is converted into a motor angle by the number of pole pairs p, and the rotor speed is estimated by integration.

[0028] An observation system based on the aforementioned observation method includes:

[0029] Stator voltage and stator current sampling module: Acquires the three-phase voltage and current signals of the motor stator, and obtains the values ​​in the stationary coordinate system via Clarke transformation. , ;

[0030] Nonlinear flux linkage observation module: Real-time estimation of flux linkage vector based on motor model;

[0031] Magnetic flux normalization module: performs amplitude normalization processing on the magnetic flux vector;

[0032] Phase-locked loop (PLL): Constructs an orthogonal error signal based on the magnetic flux phase, and obtains the estimated angle and speed of the motor through a PIR controller.

[0033] An apparatus includes at least one processor and a memory communicatively connected to the at least one processor, wherein the memory stores instructions executable by the at least one processor to enable the at least one processor to perform the observation method.

[0034] A computer-readable storage medium storing computer-executable instructions for causing a computer to perform the observation method.

[0035] Compared with the prior art, the present invention has the following advantages and beneficial effects:

[0036] This invention effectively solves the negative impact of periodic disturbances caused by periodic fluctuations in DC bus voltage on rotor position observation accuracy and system stability in traditional electrolytic capacitor-free drive systems. It significantly improves the performance of motor systems controlled without position sensors in ECFPDS and breaks through the industry bottlenecks caused by existing reliance on electrolytic capacitors, such as large size, short lifespan, and high cost.

[0037] This invention is the first to directly introduce the concept of resonant control into the phase-locked loop structure of the ECFPDS flux linkage observer. It can actively suppress disturbance signals caused by the power grid frequency and its integer multiple harmonic components, significantly improving the accuracy of sensorless observation and rotor position estimation. It overcomes the drawbacks of existing technologies that mainly rely on filtering, such as delay and slow dynamic response.

[0038] This invention has a simple structure, can be directly implemented in discrete form, is highly compatible with mainstream control strategies, and is easy to maintain and integrate while ensuring high performance, which is beneficial for engineering implementation and later maintenance.

[0039] This invention demonstrates stronger robustness of the observer under conditions such as gradient search gain changes and parameter disturbances, effectively improving the dynamic and steady-state performance of the control system and meeting the stringent response and accuracy requirements of high-performance drive systems.

[0040] This invention is applicable to both single-phase and three-phase ECFPDS, exhibiting high adaptability, low system dependence, and ease of porting to different systems. Attached Figure Description

[0041] Figure 1 This is a schematic diagram of the structure of the present invention.

[0042] Figure 2 This is a schematic diagram of the structure of the ECFPD motor drive system used in this embodiment. Detailed Implementation

[0043] The present invention will be further described in detail below with reference to the embodiments, but the implementation of the present invention is not limited thereto.

[0044] Example

[0045] like Figure 1 As shown, an improved flux linkage observer without position sensors is implemented based on the following ECFPD motor drive system. In this embodiment, the system consists of outer speed control, inner current control, space vector modulation, and sensorless position estimation. Specifically, the outer speed control uses a speed loop controller, and the inner current control uses a q-axis current controller and a d-axis current controller. The space vector modulation and sensorless position estimation include two parts: a flux linkage observer based on an improved gradient descent method and a phase-locked loop based on PIR.Figure 2 The part shown in the red box with the dashed line is the focus of this embodiment.

[0046] The working process of the ECFPD motor drive system is as follows:

[0047] The speed loop controller operates based on a given electrical angular velocity. And estimating electric angular velocity The difference generates a given value for the current loop. The dq axis current loop controller adjusts i respectively d i q， The stator voltage is generated through Park inverse transformation and SVPWM (space vector pulse width modulation) module. , The current from the three-phase inverter is output after passing through the Clark converter. , The , Input the Park transform and the flux linkage observer based on the improved gradient descent method, respectively;

[0048] The flux linkage observer based on the improved gradient descent method is specifically: based on an improved nonlinear gradient descent algorithm, the sampled... , Real-time integral estimation is performed to obtain the stator flux linkage. This observer adds a phase correction component to the traditional amplitude correction, ensuring the flux estimation remains stable during bus fluctuations. The output signal, after amplitude normalization and quadrature error calculation, provides input to the phase-locked loop.

[0049] The orthogonality error calculation specifically involves calculating the observed flux linkage vector and the phase-locked reference vector. Orthogonality error:

[0050]

[0051] A dynamic error signal reflecting the phase difference of the magnetic flux linkage is obtained.

[0052] The phase-locked loop uses a PIR controller:

[0053]

[0054] The closed-loop working principle of the motor drive system of this invention is as follows: The entire system constitutes a complete sensorless closed-loop control from "bus voltage – stator current – ​​flux linkage observation – angle estimation – speed and current control – PWM modulation". The module within the red dashed box forms a closed-loop structure of "disturbance sensing – phase correction – resonance suppression – angle feedback", which is the key part that distinguishes this invention from the prior art.

[0055] The sensorless observation method for the improved magnetic flux observer of the present invention, based on the above system, includes the following steps:

[0056] Construct a motor model and estimate the flux linkage vector, specifically as follows:

[0057] The three-phase current of the motor is obtained in a two-phase stationary coordinate system by performing Clark transform. ;

[0058] Perform an inverse Park transformation on the output of the current controller to calculate the voltage in the two-phase stationary coordinate system. ;

[0059] In a stationary coordinate system, the stator voltage, stator current, and flux linkage vector satisfy the following:

[0060]

[0061]

[0062] in, For stator flux linkage, Let Ls be the stator flux linkage differential, and Ls and Rs be the inductance and resistance of the motor, respectively.

[0063] The flux linkage vector is normalized to construct an orthogonal error signal.

[0064] Since the flux linkage amplitude varies with factors such as speed and current, directly inputting it into the phase-locked loop will lead to an increase in phase detection error. To ensure accurate phase detection, the observed flux linkage signal is normalized and an orthogonal error signal is constructed.

[0065] Orthogonal error signal: Calculations on the signal reveal that the error signal contains disturbance signals that are integer multiples of the grid frequency, caused by fluctuations in the DC bus voltage.

[0066] The traditional PI controller is extended into a PIR controller structure with a resonant element.

[0067]

[0068] If it is a single-phase ECFPDS, then If it is a three-phase ECFPDS, then ; The damping coefficient; This is the resonant gain.

[0069] The observation system in this embodiment, according to the different functions of each part, includes the following parts:

[0070] Stator voltage and stator current sampling module: Acquires the three-phase voltage and current signals of the motor stator, and obtains the values ​​in the stationary coordinate system via Clarke transformation. , ;

[0071] Nonlinear flux linkage observation module: Real-time estimation of flux linkage vector based on motor model;

[0072] Magnetic flux normalization module: performs amplitude normalization processing on the magnetic flux vector;

[0073] Phase-locked loop (PLL): Constructs an orthogonal error signal based on the magnetic flux phase, and obtains the estimated angle and speed of the motor through a PIR controller.

[0074] This embodiment also provides an apparatus including at least one processor and a memory communicatively connected to the at least one processor, wherein the memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to enable the at least one processor to perform the observation method.

[0075] This embodiment also provides a computer-readable storage medium storing computer-executable instructions for causing a computer to perform the observation method.

[0076] The above embodiments are preferred embodiments of the present invention, but the embodiments of the present invention are not limited to the embodiments described above. Any changes, modifications, substitutions, combinations, or simplifications made without departing from the spirit and principle of the present invention shall be considered equivalent substitutions and shall be included within the protection scope of the present invention.

Claims

1. An improved magnetic flux observation method without position sensors, characterized in that, Includes the following steps: Construct a motor model and estimate the flux linkage vector; The flux linkage vector is normalized to construct an orthogonal error signal. Input the phase-locked loop to obtain the motor angle and the estimated motor speed; Specifically, the estimated flux linkage vector is estimated by observing the flux linkage vector using an improved nonlinear observation law. The improved nonlinear observation law is specifically as follows: in, , Gradient gain; , The phase correction symbol has the following discrete form: , For rotor flux, For permanent magnet flux linkage; The phase-locked loop is specifically a PIR controller with a resonant element; The structure of the PIR controller with resonant loop is as follows: in, If it is a single-phase ECFPDS, then If it is a three-phase ECFPDS, then ; The damping coefficient; For resonant gain, , These are the proportional coefficient and the integral coefficient.

2. The observation method according to claim 1, characterized in that, The specific steps for constructing the motor model are as follows: The stator current in the two-phase stationary coordinate system is obtained by performing Clark transform on the three-phase current of the motor. ; Perform an inverse Park transformation on the output of the current controller to calculate the stator voltage in the two-phase stationary coordinate system. ; In a stationary coordinate system, the stator voltage, stator current, and flux linkage vector satisfy the following: in, For stator flux linkage, Let Ls be the stator flux linkage differential, and Ls and Rs be the inductance and resistance of the motor, respectively.

3. The observation method according to claim 1, characterized in that, Inputting the phase-locked loop (PLL) yields the motor angle and estimated rotor speed, specifically: The phase-locked loop outputs a mechanical angular velocity, which is converted into a motor angle by the number of pole pairs p, and the rotor speed is estimated by integration.

4. A system based on the observation method according to any one of claims 1-3, characterized in that, include: Stator voltage and stator current sampling module: Acquires the three-phase voltage and current signals of the motor stator, and obtains the values ​​in the stationary coordinate system via Clarke transformation. , ; Nonlinear flux linkage observation module: Real-time estimation of flux linkage vector based on motor model; Magnetic flux normalization module: performs amplitude normalization processing on the magnetic flux vector; Phase-locked loop (PLL): Constructs an orthogonal error signal based on the magnetic flux phase, and obtains the estimated angle and speed of the motor through a PIR controller.

5. A device, characterized in that, It includes at least one processor and a memory communicatively connected to the at least one processor, wherein the memory stores instructions executable by the at least one processor to enable the at least one processor to perform the observation method as described in any one of claims 1-4.

6. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer-executable instructions for causing a computer to perform the observation method as described in any one of claims 1-4.

Images