阅读说明：本技术 一种基于分段阈值的永磁同步电机无位置传感器控制方法 (Permanent magnet synchronous motor position sensorless control method based on segmentation threshold ) 是由 张荣标 杨铭刚 张业成 于 2021-07-13 设计创作，主要内容包括：本发明公开电机控制领域中的一种基于分段阈值的永磁同步电机无位置传感器控制方法,对当前速度进行分段处理,结合加速度和转换时间,判断出永磁同步电机的运行状态,在静止启动或低速运行状态时采用高频方波电压信号注入法,在中速或高速运行状态时采用限幅平均滤波观测法,通过将高频方波电压信号注入法与限幅平均滤波观测法相结合,不仅解决在静止启动、低速时反电动势法无法估算的问题,还解决中高速时高频方波电压信号注入法误差大的问题；采用傅里叶分解方法,将高频电流响应分解为正弦基波及高次谐波,有效地减小采样延迟等因素的影响；通过定子电流的误差来重构反电动势,采用限幅平均滤波代替传统的低通滤波法,减少相位滞后。(The invention discloses a permanent magnet synchronous motor position sensorless control method based on a segmented threshold value in the field of motor control, which comprises the steps of conducting segmented processing on the current speed, judging the running state of a permanent magnet synchronous motor by combining acceleration and conversion time, adopting a high-frequency square wave voltage signal injection method in the static starting or low-speed running state, adopting an amplitude-limiting average filtering observation method in the medium-speed or high-speed running state, and combining the high-frequency square wave voltage signal injection method with the amplitude-limiting average filtering observation method, so that the problem that a counter electromotive force method cannot estimate in the static starting and low-speed state is solved, and the problem that the error of the high-frequency square wave voltage signal injection method is large in the medium-high speed state is also solved; a Fourier decomposition method is adopted to decompose the high-frequency current response into sine fundamental waves and higher harmonics, so that the influence of factors such as sampling delay and the like is effectively reduced; the back electromotive force is reconstructed through the error of the stator current, and the traditional low-pass filtering method is replaced by amplitude limiting average filtering, so that the phase lag is reduced.)

1. A permanent magnet synchronous motor position sensorless control method based on a segmentation threshold value is characterized by comprising the following steps:

step 1) acquiring and calculating the current speed V and the acceleration a of the permanent magnet synchronous motor, and carrying out sectional processing on the current speed V, wherein V is more than or equal to 0 and less than or equal to V_m30% of the first stage when V_m*30％<V<V_{Forehead (forehead)}For the second stage, the conversion time of the two stages is calculatedV_mIs the rated speed of the permanent magnet synchronous motor;

step 2) combining the acceleration a and the transition time t_mJudging four running states of the permanent magnet synchronous motor, namely, static starting, low speed, medium speed or high speed;

step 3) when the permanent magnet synchronous motor is in a static starting or low-speed running state, controlling the permanent magnet synchronous motor by adopting a high-frequency square wave voltage signal injection method; and when the permanent magnet synchronous motor is in a medium-speed or high-speed running state, controlling the permanent magnet synchronous motor by adopting an amplitude limiting average filtering observation method.

2. The method for controlling the permanent magnet synchronous motor without the position sensor based on the segmentation threshold value as claimed in claim 1, wherein the method comprises the following steps: in step 2), when the permanent magnet synchronous motor is in the first stage, when a is 0 or a<0 or when a>0 and 0. ltoreq. t<t_mJudging whether the permanent magnet synchronous motor is in a static starting or low-speed running state; when a is>0 and t is more than or equal to t_mAnd judging the medium-speed or high-speed running state of the permanent magnet synchronous motor.

3. The method for controlling the permanent magnet synchronous motor without the position sensor based on the segmentation threshold value as claimed in claim 1, wherein the method comprises the following steps: in step 2), when the permanent magnet synchronous motor is in the second stage, when a is 0 or a>At 0 or a<0 and 0. ltoreq. t<t_mJudging the medium-speed or high-speed running state of the permanent magnet synchronous motor; when a is<0 and t is more than or equal to t_mAnd judging whether the permanent magnet synchronous motor is in a static starting or low-speed running state.

4. The method for controlling the permanent magnet synchronous motor without the position sensor based on the segmentation threshold value as claimed in claim 1, wherein the method comprises the following steps: in step 3), the high-frequency square wave voltage signal injection method comprises the following steps:

step (1): establishing an estimated rotor rotation coordinate system of a permanent magnet synchronous motorThe actual rotor rotating coordinate system dq and the stationary coordinate system alpha beta to the estimated rotating coordinate systemMedium injection high frequency square wave voltage Obtaining the stator voltage u after the injection of the high-frequency square wave voltage signal_dh，u_qhAnd estimating a rotating coordinate systemHigh frequency current on the shaft

Step (2): for high frequency currentPerforming Fourier decomposition, and responding the high-frequency current after Fourier decompositionMultiplying the opposite number by cos (omega x), and obtaining a position error function f (delta theta) containing position error information through low-pass filtering;

and (3): adjusting the position error function f (delta theta) to zero through PI, and outputting an estimated rotor position angleEstimating rotor position angleObtaining permanent magnet synchronous motor after differential linkEstimating velocity

5. The method for controlling the permanent magnet synchronous motor without the position sensor based on the segmentation threshold value as claimed in claim 1, wherein the method comprises the following steps: in step 3), the high-frequency square wave voltage signal injection method comprises the following steps:

step (1): by stator current errorAs a slip-form surface of the slip-form observer,observed stator currents in alpha, beta axes, i_α、i_βObtaining the extended back electromotive force e for the actual stator current of the alpha and beta axes according to the sliding mode control law_α、e_β；

Step (2): will expand the back emf e_α、e_βObtaining smooth extended back electromotive force after amplitude limiting average filtering

And (3): smooth extended back emfObtaining the estimated rotor position angle through the operation of an arc tangent functionEstimating rotor position angleObtaining the estimated speed of the permanent magnet synchronous motor after a differentiation link

6. The method for controlling the permanent magnet synchronous motor without the position sensor based on the segmentation threshold value as claimed in claim 5, wherein the method comprises the following steps: in the step (2), the amplitude-limited average filtering is as follows: firstly, respectively sampling n extended back electromotive forces e_α、e_βTaking the average of the two groups of n expanded back electromotive force values as the historical sampling average value of the two signals, and then carrying out the expansion back electromotive force signal e_α、e_βSampling for the (n +1) th time, comparing the sampling value for the (n +1) th time with the historical sampling average value, and if the difference value is smaller than the set maximum sampling deviation value, the sampling value is valid; if the difference value is larger than the set maximum sampling deviation value, the sampling value at the time is invalid, and a historical sampling average value is selected as the sampling value at the time; sending the compared sampling values into a queue for sliding-median average filtering processing, continuously selecting N values from the compared sampling values, wherein N is more than or equal to 2 and less than or equal to N, regarding the N values as a queue with fixed length, putting the tail of the queue when a new data is sampled, throwing away a primary data of the original head of the queue, then removing a maximum value and a minimum value in the queue, and finally calculating the arithmetic average value of the remaining (N-2) data to be used as the latest extended back electromotive force output, namely the smooth extended back electromotive force

7. The method for controlling the permanent magnet synchronous motor without the position sensor based on the segmentation threshold value as claimed in claim 5, wherein the method comprises the following steps: in step (3), the rotor position angle is estimated

8. The method of claim 4, wherein the method comprises a step of controlling the PMSM to have no position sensor based on a segment thresholdIs characterized in that: in the step (1), the high-frequency square wave voltageV_hIs a square wave amplitude, phi_sprIs a unit square wave function.

9. The method for controlling the permanent magnet synchronous motor without the position sensor based on the segmentation threshold value as claimed in claim 8, wherein the method comprises the following steps: in step (1), the stator voltageZ_d＝R_s+jωL_dComplex impedance on the real d-axis, Z_q＝R_s+jωL_qIs the complex impedance on the actual q-axis, j is the imaginary unit, ω is the angular frequency, R_SIs stator resistance, L_dD-axis inductance, L, of actual rotor rotation coordinate axis_qQ-axis inductance, i, for actual rotor rotation axis_dhIs the stator current of the actual d-axis, i_qhIs the actual q-axis stator current.

10. The method for controlling the permanent magnet synchronous motor without the position sensor based on the segmentation threshold value as claimed in claim 9, wherein the method comprises the following steps: in step (1), the estimated rotational coordinate systemHigh frequency current on the shaft

Technical Field

The invention relates to the technical field of motor control, in particular to a position-sensorless control method of a permanent magnet synchronous motor, which is applied to high-performance machine tool control, position control, automobile transmission and the like.

Background

The permanent magnet synchronous motor has the advantages of simple structure, high response speed, high operation efficiency, small torque pulsation and the like, and a driving system of the permanent magnet synchronous motor is composed of a motor, an inverter, a controller, a sensor and the like, wherein the operation can be influenced when any one part of the parts breaks down. Accurate capture of rotor position information is essential to reduce failures and achieve high performance field control. The permanent magnet synchronous motor generally acquires real-time rotor position information through a mounting position sensor, and vector control of the permanent magnet synchronous motor is completed. However, the position sensor has strong sensitivity to the environment and is prone to malfunction. A faulty position sensor can cause the motor to operate unstably or even out of control. It is important to replace the position sensor with a position sensorless control in the event of a fault condition.

The document with the Chinese patent application number of 201911063202.X discloses a position-sensorless control method of a permanent magnet synchronous motor in a medium-high speed state, which mainly utilizes a sliding mode control theory to estimate a back electromotive force value and perform position estimation on a rotor of the permanent magnet synchronous motor so as to complete position-sensorless control. However, this control method cannot be used when the permanent magnet synchronous motor is in a stationary start or at a low speed because the back electromotive force is too small to be estimated. To solve this problem, chinese patent application No. 201610954686.7 discloses a method for controlling a permanent magnet synchronous motor without a position sensor in a low speed state, which mainly uses a signal superposition principle to perform position detection by injecting a high-frequency sinusoidal voltage signal into the motor, thereby completing the control of the permanent magnet synchronous motor without a position sensor in a low speed state. However, the method has a high requirement on the accuracy of the sampling point, when the sampling point is influenced and shifted, an error will be generated, and the differential current is calculated, so that the influence of factors such as sampling delay and the like can be caused. The chinese patent application No. 201410625883.5 discloses a sensorless control method for a permanent magnet synchronous motor in a medium-high speed state, which mainly uses a current angle detection method to determine a real-time position of a rotor of the permanent magnet synchronous motor by detecting a current angle error of a rotating shaft, thereby completing sensorless control of the permanent magnet synchronous motor. However, the method needs low-pass filtering and has a certain phase lag, the obtained result needs compensation, the process is complex, and the response speed of the system operation is slow.

Disclosure of Invention

The invention aims to solve the problems of the existing permanent magnet synchronous motor sensorless control technology, and provides a sensorless control method based on a segmentation threshold value for a permanent magnet synchronous motor at any speed. When the permanent magnet synchronous motor is started statically or at a low speed, a high-frequency square wave voltage signal injection method without calculating differential current is adopted for position-sensor-free control; when the permanent magnet synchronous motor is in a high speed state, a position-sensorless control is carried out by adopting an amplitude limiting average filtering observation method without angle compensation.

In order to achieve the purpose, the technical scheme adopted by the permanent magnet synchronous motor position sensorless control method based on the segmented threshold value comprises the following steps:

step 1) acquiring and calculating the current speed V and the acceleration a of the permanent magnet synchronous motor, and carrying out sectional processing on the current speed V, wherein V is more than or equal to 0 and less than or equal to V_m30% of the first stage when V_m*30％<V<V_{Forehead (forehead)}For the second stage, the conversion time of the two stages is calculatedV_mIs the rated speed of the permanent magnet synchronous motor;

step 2) combining the acceleration a and the transition time t_mJudging four running states of the permanent magnet synchronous motor, namely, static starting, low speed, medium speed or high speed;

step 3) when the permanent magnet synchronous motor is in a static starting or low-speed running state, controlling the permanent magnet synchronous motor by adopting a high-frequency square wave voltage signal injection method; and when the permanent magnet synchronous motor is in a medium-speed or high-speed running state, controlling the permanent magnet synchronous motor by adopting an amplitude limiting average filtering observation method.

Further, in step 2), when the permanent magnet synchronous motor is in the first stage, when a is 0 or a<0 or when a>0 and 0. ltoreq. t<t_mJudging whether the permanent magnet synchronous motor is in a static starting or low-speed running state; when a is>0 and t is more than or equal to t_mAnd judging the medium-speed or high-speed running state of the permanent magnet synchronous motor.

Further, in step 2), when the permanent magnet synchronous motor is in the second stage, when a is 0 or a>At 0 or a<0 and 0. ltoreq. t<t_mJudging the medium-speed or high-speed running state of the permanent magnet synchronous motor; when a is<0 and t is more than or equal to t_mAnd judging whether the permanent magnet synchronous motor is in a static starting or low-speed running state.

Further, in step 3), the high-frequency square wave voltage signal injection method is:

step (1): establishing an estimated rotor rotation coordinate system of a permanent magnet synchronous motorThe actual rotor rotating coordinate system dq and the stationary coordinate system alpha beta to the estimated rotating coordinate systemMedium injection high frequency square wave voltageObtaining the stator voltage u after the injection of the high-frequency square wave voltage signal_dh，u_qhAnd estimating a rotating coordinate systemHigh frequency current on the shaft

Step (2): for high frequency currentPerforming Fourier decomposition, and responding the high-frequency current after Fourier decompositionMultiplying the opposite number by cos (omega x), and obtaining a position error function f (delta theta) containing position error information through low-pass filtering;

and (3): adjusting the position error function f (delta theta) to zero through PI, and outputting an estimated rotor position angleEstimating rotor position angleObtaining the estimated speed of the permanent magnet synchronous motor after a differentiation link

Further, in step 3), the high-frequency square wave voltage signal injection method is:

step (1): by stator current errorAs a slip-form surface of the slip-form observer,observed stator currents in alpha, beta axes, i_α、i_βObtaining the extended back electromotive force e for the actual stator current of the alpha and beta axes according to the sliding mode control law_α、e_β；

Step (2): will expand the back emf e_α、e_βObtaining smooth extended back electromotive force after amplitude limiting average filtering

And (3):smooth extended back emfObtaining the estimated rotor position angle through the operation of an arc tangent functionEstimating rotor position angleObtaining the estimated speed of the permanent magnet synchronous motor after a differentiation link

Compared with the prior art, the invention has the following advantages:

1. aiming at the problem that the control of the permanent magnet synchronous motor without the position sensor is only suitable for a specific speed range at present, the permanent magnet synchronous motor can be stably switched among static start, low speed and medium and high speed by segmenting the running state of the motor based on speed and acceleration, the running state of the permanent magnet synchronous motor is segmented and then subjected to fault tolerance, and the problem of the speed range of the permanent magnet synchronous motor without the position sensor is effectively solved.

2. Aiming at the problem of nonlinear influence caused by calculation of differential current in the existing high-frequency square wave signal injection method, the invention adopts a Fourier decomposition method to decompose high-frequency current response into sine fundamental waves and higher harmonics, thereby effectively reducing the influence of factors such as sampling delay and the like, obtaining rotor position information by using the transformation of continuous functions, avoiding calculating differential current, effectively solving the influence of nonlinear factors such as sampling delay and the like caused by high sampling frequency, increasing the bandwidth of a control system without a position sensor and improving the dynamic response speed.

3. According to the amplitude limiting average filtering observation method, the counter electromotive force is reconstructed through the error of the stator current, the traditional low-pass filtering method is replaced by the amplitude limiting average filtering, the phase lag is reduced, and the problems that the phase lag needs to be compensated and the system response is slow are solved, so that the rapidity and the control precision of the permanent magnet synchronous motor control without the position sensor are improved.

4. According to the invention, the high-frequency square wave voltage signal injection method is combined with the amplitude-limiting average filtering observation method, so that the problem that the back electromotive force method cannot estimate the permanent magnet synchronous motor at the static starting and low speed is solved, the problem that the high-frequency square wave voltage signal injection method has large error at the medium-high speed is also solved, and the flexibility of the permanent magnet synchronous motor in the control without a position sensor is effectively improved.

Drawings

FIG. 1 is a schematic diagram of a control system connection for implementing the control method of the present invention;

FIG. 2 is a flow chart of a method for sensorless control of a PMSM based on segment threshold in accordance with the present invention;

FIG. 3 is a control flow diagram of the sectional processing of the permanent magnet synchronous motor speed until the high frequency square wave voltage signal injection method and the clipping average filtering observation method are selected in FIG. 2;

FIG. 4 is a diagram illustrating a positional relationship between the estimated rotor synchronous rotation coordinate system and the actual rotor synchronous rotation coordinate system when the high-frequency square wave voltage signal injection method is employed in FIG. 2;

FIG. 5 is a diagram illustrating the rotation coordinate system extracted and estimated in FIG. 2 by the high-frequency square-wave voltage signal injection methodAngle schematic in shaft high frequency current;

FIG. 6 is a schematic diagram of the clipped average filtering observation employed in FIG. 2;

the serial numbers and designations of the various components in the drawings: 1. a permanent magnet synchronous motor; 2. a drive module; 3. a DSP28335 controller; 4. a processor; 5. a programmable counter.

Detailed Description

Referring to fig. 1, a control system implementing the present invention is connected to a permanent magnet synchronous motor 1, the control system comprising a drive module 2, a DSP28335 controller 3, a processor 4 and a programmable counter 5. The drive module 2 is in bidirectional connection with the DSP28335 controller 3 to realize communication, 220V alternating current is input into the permanent magnet synchronous motor 1 after being rectified through the drive module 2, and the DSP28335 controller 3 sends a switching tube control signal to drive the permanent magnet synchronous motor 1. The programmable counter 5 collects the rotating speed of the permanent magnet synchronous motor 1, the output end of the programmable counter 5 is connected with the processor 4, a rotating speed pulse signal in the programmable counter 5 is transmitted to the processor 4 through a lead, the processor 4 is an Stm32 processor, the processor 4 processes the rotating speed pulse signal, the current speed V and the acceleration a of the permanent magnet synchronous motor 1 are calculated through rotating speed information, the current speed V is processed in a segmented mode, and the running state of the motor is obtained by combining the acceleration a: namely four operating states of stationary start, low speed, medium speed or high speed. Aiming at a medium-speed or high-speed running state, outputting a corresponding group of high levels '11'; and outputting a corresponding group of low-level 00 signals aiming at the static starting or low-speed running state. The output end of the processor 4 is connected with the DSP28335 controller 3, and the processor 4 sends a group of high level '11' or a group of low level '00' signals to the DSP28335 controller 3. The DSP28335 controller 3 selects different control methods according to the transmitted level information: if the signal is a low-level '00' signal, a high-frequency square wave voltage signal injection method is adopted, and the estimated speed obtained by the high-frequency square wave voltage signal injection method is sent to the driving module 2; if the high level is '11', an amplitude limiting average filtering observation method is adopted, and the estimated speed obtained by the amplitude limiting average filtering observation method is sent to the driving module 2, so that the permanent magnet synchronous motor 1 is controlled.

Referring to fig. 2 and 3, when the processor 4 performs the segmentation processing on the current speed V, the change of the running state of the permanent magnet synchronous motor 1 is smoothly transited by integrating the speed and the acceleration. Firstly, judging the acceleration state of the permanent magnet synchronous motor according to the positive and negative information of the acceleration a: if a>0, the permanent magnet synchronous motor is in an accelerated running state; if a is 0, the permanent magnet synchronous motor is in a state of constant-speed operation; if a<And 0, the permanent magnet synchronous motor is in a speed reduction running state. Then, according to the speed manual of the permanent magnet synchronous motor, the rated speed V of the permanent magnet synchronous motor is used_m30% of the reference standard, when the current speed V of the permanent magnet synchronous motor is greater than the rated speed V_m30% of the total motor heightOperating in a fast state; when the current speed V of the permanent magnet synchronous motor is less than or equal to the rated speed V_mWhen the speed is 30%, the motor runs in a low-speed state; when the current speed of the permanent magnet synchronous motor is 0, the motor is in a static starting state. Therefore, the running state of the permanent magnet synchronous motor can be divided into two stages according to the current speed V of the permanent magnet synchronous motor: the current speed in the first stage is: v is more than or equal to 0 and less than or equal to V_m30% of the total weight; the current speed of the second stage is: v_m*30％<V<V_{Forehead (forehead)}Calculating the conversion time t of two stages when the permanent magnet synchronous motor runs_m：

Referring to fig. 2 and 3, when the processor 4 determines that it is operating in the first phase according to the current speed V, and when a is equal to 0, it indicates that the permanent magnet synchronous motor is operating at a constant speed, and does not enter the second phase, and when a is equal to 0<When the speed is 0, judging that the permanent magnet synchronous motor operates at a reduced speed, and the second stage can not be started; when a is>When 0, judging that the permanent magnet synchronous motor runs in an accelerated way, and then, converting for time t_mThen the second phase is entered. Therefore, when the permanent magnet synchronous motor is in the first stage, and a is 0 or a<When 0, judging that the motor is in a static starting or low-speed running state; when a is>0, recombination transition time t_mAnd when 0. ltoreq. t<t_mAnd judging that the permanent magnet synchronous motor is in a static starting or low-speed running state, wherein t is the current running time of the permanent magnet synchronous motor. When the permanent magnet synchronous motor is in a static starting state or a low-speed running state, the processor 4 sends low level '00' to the DSP28335 controller 3, and the DSP28335 controller 3 adopts a high-frequency square wave voltage signal injection method. When a is>0, and t is not less than t_mWhen the motor is in a medium-speed or high-speed running state, the processor 4 sends a high-level '11' signal to the DSP28335 controller 3, and the DSP28335 controller 3 adopts a limiting average filtering observation method.

When the processor 4 judges that the permanent magnet synchronous motor operates in the second stage according to the current speed V, and when a is equal to 0, the permanent magnet synchronous motor is judged to operate at a constant speedThe first stage is not entered; when a is>When the speed is 0, judging that the permanent magnet synchronous motor runs in an accelerated mode, and the first stage cannot be started; when a is<When 0, judging that the permanent magnet synchronous motor operates in a speed reduction mode, and then, converting for time t_mThen the first stage is entered. When the permanent magnet synchronous motor is in the second stage, when a is 0 or a>When 0, judging that the device is in a medium-speed or high-speed running state; when a is<0 and 0. ltoreq.t<t_mWhen the permanent magnet synchronous motor is in a medium-high speed running state, the processor 4 sends a high level '11' signal to the DSP28335 controller 3, and the DSP28335 controller 3 adopts a limiting average filtering observation method. When the permanent magnet synchronous motor is in the second stage, when a<0, and t is more than or equal to t_mAnd at the moment, the processor 4 sends low level '00' to the DSP28335 controller 3, and the DSP28335 controller 3 adopts a high-frequency square wave voltage signal injection method.

Referring to fig. 4, when the high frequency square wave voltage signal injection method is used, the DSP28335 control board 3 establishes an estimated rotor rotation coordinate system of the permanent magnet synchronous motorAn actual rotor rotational coordinate system dq and a stationary coordinate system α β, wherein the rotor rotational coordinate system is estimatedAt an angle to the stationary coordinate system α β ofTo estimate the rotor position angle. The angle between the actual rotor rotational coordinate system dq and the stationary coordinate system α β is θ, which is the actual rotor position angle. Estimating a rotor rotation coordinate systemThe angle between the rotor and the actual rotor rotating coordinate system dq is delta theta, the delta theta is the estimated error angle of the rotor, and then the estimated error angle of the rotor is obtainedThe angle theta between the actual rotor rotating coordinate system dq and the static coordinate system alpha beta of the permanent magnet synchronous motor accurately reflects the running position of the permanent magnet synchronous motor, but cannot be solved directly, so that the estimated rotor position angle can be solved by adjusting the estimated rotor error angle delta theta to be 0Thereby indirectly obtaining θ.

To adjust the rotor estimation error angle Δ θ, the estimated rotational coordinate system is adjustedInjecting a high-frequency square wave voltage signal, wherein the signal is expressed as follows:

in the formula (I), the compound is shown in the specification,for estimating a rotating coordinate systemMedium injected high frequency square wave voltage, V_hIs the square wave amplitude, phi s_prIs a unit square wave function and has the expression ofT is the square wave period, x is the injection square wave time, u (T) is the time of the injection square wave time x in the square wave period T.

According to the parameters of the permanent magnet synchronous motor: stator resistance R_SD-axis inductance L of actual rotor rotation coordinate axis_dQ-axis inductance L of actual rotor rotation coordinate axis_qObtaining the stator voltage after injecting the high-frequency square wave voltage signal as follows:

in the formula, Z_d＝R_s+jωL_dComplex impedance on the real d-axis, Z_q＝R_s+jωL_qIs the complex impedance on the actual q-axis, j is the imaginary unit, ω is the angular frequency, u_dhIs the actual d-axis stator voltage, u_qhFor stator voltage of true q-axis, i_dhIs the stator current of the actual d-axis, i_qhIs the actual q-axis stator current.

Bonding ofAnd equations (1) and (2) to obtain an estimated rotational coordinate systemThe high frequency current on the shaft is:

wherein For estimating a rotating coordinate systemHigh frequency current on the shaft, V_hFor injecting the amplitude of the square wave, φ s_prIs a unit square wave function.

Referring to FIG. 5, the rotating coordinate system is estimated for extractionHigh frequency current on the shaftSchematic diagram of the angle in (1). DSP28335 controller 3 firstly processes high-frequency currentPerforming Fourier decomposition, and obtaining the following result after applying the Fourier decomposition:

whereinIs constant, Δ Z_kJ (2k +1) ω Δ L is the difference between the complex impedances of the actual d and q axes, Z_dk＝R_s+j(2k+1)ωL_dIs the complex impedance of the actual d-axis, Z_qk＝R_s+j(2k+1)ωL_qIs the complex impedance of the actual q-axis,is the phase angle of the actual d-axis,is the phase angle of the actual q axis, ω is the injected square wave frequency, Δ L is the difference between the actual d and q axis inductances, k is 0,1, 2.

Taking Fourier decomposed high frequency current responseMultiplying the inverse number by cos (ω x), and obtaining a position error function f (Δ θ) containing position error information through a low-pass filter, wherein sin (2 Δ θ) ≈ 2 Δ θ when Δ θ is small:

in the formula, k₁Is a constant other than 0, M is a constant other than 0, LPF is a low pass filter, Z_d0Is the complex impedance on the actual d-axis at low frequencies, Z_q0Is the complex impedance on the actual q-axis at low frequency, phi_d0Is the actual d-axis at low frequencyPhase angle of phi_q0Is the phase angle on the actual q-axis at low frequencies.

Then, the position error function f (delta theta) is adjusted to zero through a PI regulator, and at the moment, the output of the PI regulator is the estimated rotor position angle Will estimate the rotor position angleObtaining the estimated speed of the permanent magnet synchronous motor after a differentiation linkWill estimate the velocityThe information is transmitted to the driving module 2, and the driving module 2 estimates the speedThe corresponding current information is transmitted to the permanent magnet synchronous motor, thereby completing the position sensorless control of the permanent magnet synchronous motor 1.

See fig. 6 for a schematic diagram of a clipped average filtering observation method. The DSP28335 controller 3 writes the voltage state equation of the PMSM into a current state equation as shown in formula (6) according to the stator current model of the PMSM, and uses the current state equation as a reference model of the sliding-mode observer as shown in formula (7). i.e. i_α、i_βAnd u_α、u_βActual stator current and stator voltage, e, of the alpha and beta axes, respectively_αAnd e_βRespectively, the extended back electromotive force of alpha and beta axes, B is the matrix coefficient, L_dIs an inductance of the permanent magnet synchronous motor,and v_α、v_βRespectively being viewed on alpha and beta axesMeasuring the stator current and observing the extended back emf,

the observed stator current model is differenced with the actual stator current model to obtain a stator current error equation shown as a formula (8), and the stator current error is subtractedAs a sliding surface of the sliding observer.And v_α、v_βObserved stator currents and observed extended back emf for the α, β axes, respectively:

by correcting the designed sliding mode observer model in real time, the error between the stator current of the observer and the actual stator current is 0, the extended back electromotive force of the observer is equal to the extended back electromotive force of the actual motor, and therefore the sliding mode control law of the state observer shown in the formula (8) is designed. Where k is a constant other than 0 and Sgn is a sign function.

Finally, obtaining the extended back electromotive force e according to the sliding mode control law_α、e_βSignal, will expand the back electromotive force e_α、e_βObtaining smooth extended back electromotive force after amplitude limiting average filtering(Andfiltered extended back emf for the alpha and beta axes, respectively), thereby improving the accuracy of the rotor position estimation.

The specific method of the amplitude limiting average filtering is as follows: the DSP28335 controller 3 samples two groups of n extended back electromotive force signals e respectively_α、e_βAnd n is a natural number which is more than or equal to 1, wherein n is 1,2,3 and 4, the values of the two groups of n expanded back electromotive forces are respectively averaged to be used as a historical sampling average value of the two signals, and after a maximum sampling deviation value of the system is set, the DSP28335 control board 3 controls the expanded back electromotive force signals e_α、e_βThe (n +1) th sampling is performed. Comparing the (n +1) th sampling value with the historical sampling average value, and if the difference is smaller than the maximum sampling deviation value set by the system, the sampling value is valid; and if the difference is larger than the maximum sampling deviation value set by the system, the sampling value is invalid, the sampling value is discarded, and the historical sampling average value is selected as the sampling value. Sending the compared sampling values into a queue for sliding-median average filtering processing, continuously selecting N (N is more than or equal to 2 and less than or equal to N) values from the compared sampling values, regarding the N values as a queue with fixed length, putting the tail of the queue when a new data is sampled again, and throwing away a primary data of the original head of the queue. Then, a maximum value and a minimum value in the queue are removed, and the arithmetic mean value of the rest (N-2) data is calculated and is used as the latest extended back electromotive force signal to be output. The cut-off frequency can be reduced due to the multiple use of the sliding-median average filtering, namely, high-frequency signals with relatively high frequency can be filtered out, and the lag on the phase can not be caused. And secondary sliding-median average filtering is adopted, and circulation is performed on the basis of the primary sliding-median average filtering. Finally, smooth extended back electromotive force is obtained

Then smooth extended back electromotive force is appliedObtaining the estimated rotor position angle through the operation of an arc tangent function as shown in a formula (9)

Finally, the rotor angle will be estimatedObtaining the estimated speed of the permanent magnet synchronous motor after a differentiation linkWill estimate the velocityThe information is transmitted to the driving module 2, and the driving module 2 estimates the speedAnd transmitting the corresponding current information to the permanent magnet synchronous motor, thereby completing the position sensorless control of the permanent magnet synchronous motor.