Rotor position detection method based on pulsating high-frequency voltage injection method

Abstract

The invention discloses a pulsating high-frequency voltage injection method-based rotor position detection method, which specifically comprises the following steps of: injecting a high-frequency voltage signal, establishing a mathematical model of a permanent magnet synchronous linear motor under high-frequency excitation, extracting a high-frequency response current from # imgabs0 # shaft current by utilizing an improved quasi-proportional resonance controller, multiplying the high-frequency response current with a sinusoidal signal for modulation, and finally obtaining the position of the permanent magnet synchronous linear motor. The other improved quasi-proportional resonance controller is used for extracting a signal component with the frequency of 2wh, and a rotor position error signal is obtained through subtraction; and taking the rotor position error signal as the input of a position observer proportional integrator to obtain an estimated rotor angular speed, and integrating the estimated rotor angular speed to obtain an estimated rotor position. A high-frequency response signal is accurately locked by utilizing the gain characteristic of the improved quasi-proportional resonance controller at the resonant frequency, the dynamic delay is reduced, and the rotor position tracking precision is improved; meanwhile, through collaborative design of the resonant frequency and the bandwidth, the parameter setting process is simplified, and parameter adjustment is reduced.

Description

Technical Field

[0001] The invention belongs to the technical field of permanent magnet synchronous linear motor control, and in particular relates to a rotor position detection method based on a pulse high frequency voltage injection method. Background Art

[0002] Permanent magnet synchronous linear motor can directly drive the load to make linear motion without the need for intermediate mechanical conversion and transmission chain links, which greatly improves production efficiency. It has the advantages of simple structure, reliable operation, low mechanical consumption, low noise, good environmental adaptability, etc. It is widely used in high-precision CNC machine tools, photolithography machines, high-speed logistics and ropeless elevators. In the high-performance servo control system driven by PMLSM, in order to achieve precise position and speed control of the system, displacement / speed sensors such as grating rulers are often used to feedback the position / speed of the mover. Applying advanced non-sensing algorithms to the PMLSM servo control system can achieve high-performance control accuracy with sensing without sensors.

[0003] During the zero-speed and low-speed operation stages of the motor, the pulse high-frequency voltage injection method can track the rotor position very well. This method does not rely on the fundamental wave equation of the motor, is insensitive to changes in motor parameters, and has good robustness. However, the actual application of this method is often not as ideal as the theoretical derivation. The traditional pulse high-frequency voltage injection method uses bandpass and low-pass filters in the signal processing process to limit the bandwidth of the current loop and position observer, affecting the dynamic observation accuracy of the system. In actual system debugging, the parameter setting of the filter also makes the parameter adjustment of the system regulator complicated. Summary of the invention

[0004] The object of the present invention is to provide a rotor position detection method based on a pulse high-frequency voltage injection method, which reduces dynamic delay and improves the tracking accuracy of the rotor position.

[0005] The technical solution adopted by the present invention is a rotor position detection method based on a pulse high-frequency voltage injection method, which is specifically implemented according to the following steps:

[0006] Step 1, inject a high-frequency voltage signal into the d-axis, establish a mathematical model of the permanent magnet synchronous linear motor under high-frequency excitation, and then obtain the relationship between the high-frequency voltage signal and the high-frequency current signal in the estimated coordinate system, and finally obtain the current response equation in the rotating coordinate system;

[0007] Step 2: Using the improved quasi-proportional resonant controller Shaft current Extract high frequency response current from Then with the sinusoidal signal sin(w h t) multiplied by modulation, and another improved quasi-proportional resonant controller was used to extract a frequency of 2w hThe signal component is then subtracted once to obtain the rotor position error signal f(Δθ);

[0008] Step 3: Use the rotor position error signal f(Δθ) as the input of the position observer proportional integrator to obtain the estimated rotor angular velocity, and integrate the estimated rotor angular velocity to obtain the estimated rotor position.

[0009] The present invention is also characterized in that:

[0010] In step 1, specifically:

[0011] Under high-frequency sinusoidal excitation, the high-frequency mathematical model of the permanent magnet synchronous linear motor is obtained as shown in the following formula (2):

[0012]

[0013] In the formula, u dh 、u qh is the d-axis high frequency response voltage and the q-axis high frequency response voltage; i dh 、i qh is the d-axis high frequency response current and the q-axis high frequency response current; L dh , L qh are the d-axis high frequency response inductance and the q-axis high frequency response inductance;

[0014] According to the high-frequency mathematical model, the current response equation in the actual coordinate system can be obtained, as shown in formula (3):

[0015]

[0016] The real coordinate system of d and q axes The voltage transformation relationship and current transformation relationship in the axis estimation coordinate system are shown in equations (4) and (5);

[0017]

[0018] In the formula, express Axis high frequency response voltage, Axis high frequency response voltage; express Axis high frequency response current, express The axis high frequency response current; Δθ is the difference between the actual value θ and the estimated value of the electrical angle difference;

[0019] d, q axis real coordinate system to The axis estimation coordinate system transformation matrix is ​​shown in formula (7):

[0020]

[0021] According to formulas (3), (4), (5), and (7), the relationship between the high-frequency voltage signal and the high-frequency current signal in the estimated coordinate system is obtained, as shown in formula (8):

[0022]

[0023] Let the mean inductance be formula (9):

[0024] L=(L dh +L qh ) / 2 (9);

[0025] The differential inductance is (10)

[0026] ΔL=(L dh -L qh ) / 2 (10);

[0027] Substituting the mean inductance and differential inductance expressed by equations (9) and (10) into equation (8), we can obtain the relationship between the high-frequency voltage signal and the high-frequency current signal in the estimated coordinate system, as shown in equation (11):

[0028]

[0029] exist The shaft injects a high-frequency sinusoidal voltage, as shown in equation (12):

[0030]

[0031] Among them, u h is the amplitude of the injected high-frequency sinusoidal voltage, ω h is the angular frequency of the injected high-frequency sinusoidal voltage, and t is the time;

[0032] Substituting equation (12) into equation (11), we get the current response equation in the rotating coordinate system, as shown in equation (13):

[0033]

[0034] In step 2, As shown in formula (14):

[0035]

[0036] In the formula, K pr is the corresponding resonance parameter at high frequency, ω h is the angular frequency of the injected signal, ω c is the bandwidth cut-off frequency.

[0037] In step 2, f(Δθ) is as shown in formula (15):

[0038]

[0039] In the formula, The subscript ω h , 2ω h They respectively correspond to the frequencies of the extracted signals required to improve the quasi-proportional resonant controller.

[0040] In step 3, the rotor position error signal f(Δθ) is used as the input of the position observer proportional integrator to obtain the estimated rotor angular velocity, as shown in equation (16):

[0041]

[0042] In step 3, the estimated rotor angular velocity is integrated to obtain the estimated rotor position, as shown in equation (17):

[0043] θ e =∫ω e (17).

[0044] The beneficial effects of the present invention are as follows: the rotor position detection method based on the pulse high-frequency voltage injection method of the present invention utilizes the gain characteristics of the improved quasi-proportional resonant controller at the resonant frequency to accurately lock the high-frequency response signal, reduce dynamic delay and improve the rotor position tracking accuracy; at the same time, the improved quasi-proportional resonant controller simplifies the parameter setting process and reduces parameter adjustment through the coordinated design of the resonant frequency and bandwidth; and the method omits the traditional filtering circuit and mechanical sensor, simplifies the hardware structure and reduces the implementation cost. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] Figure 1 It is a control block diagram of a sensorless system of a permanent magnet synchronous linear motor based on rotor position detection of the present invention;

[0046] Figure 2 It is a relationship diagram of each coordinate system in the present invention;

[0047] Figure 3 This is the structure diagram of the rotor position observer based on QPR;

[0048] Figure 4 It is the Bode diagram of the improved quasi-proportional resonant controller;

[0049] Figure 5 This is the structural block diagram of the improved quasi-proportional resonant controller.

[0050] Figure 6 The actual rotor position and estimated rotor position waveform diagram of the present invention;

[0051] Figure 7 is a rotor position error waveform diagram of the present invention;

[0052] Figure 8It is the waveform diagram of actual rotation speed and estimated rotation speed of the present invention. DETAILED DESCRIPTION

[0053] The present invention is described in detail below in conjunction with specific implementation modes and accompanying drawings.

[0054] Example 1

[0055] The present invention is based on the rotor position detection method of the pulse high-frequency voltage injection method, such as Figure 1 As shown, a pulse high-frequency voltage signal is superimposed on the d-axis, and the speed and current loops both use PI regulators. The current loop uses a first-order low-pass filter to reduce the impact of the high-frequency response current signal on the fundamental current loop. The voltage output by the dq-axis current regulator is inversely transformed by Park to obtain the voltage in the two-phase static α-β coordinate system, and then the space vector pulse width modulation strategy (SVPWM) is used to obtain the six-way switch signal of the three-phase inverter, and the permanent magnet synchronous linear motor (PMLSM) is driven to detect the three-phase current of the permanent magnet synchronous linear motor. The three-phase current of the permanent magnet synchronous linear motor is Clark transformed to obtain the stator current of the permanent magnet synchronous linear motor in the α-β coordinate system. α 、i β , the current in the dq coordinate system is obtained through Park transformation. Figure 3 As shown, the q-axis current response of the estimated rotor synchronous rotating frame is extracted through an improved quasi-proportional resonant controller (QPR) with a frequency of w h The high frequency response current component is then combined with the sinusoidal signal sin(w h t) multiplied by each other to modulate and obtain the base frequency component and the frequency 2w h The high frequency component of the quasi-proportional resonant controller is used to extract the frequency of 2w h The high-frequency signal is finally used to subtract the frequency of 2w from the modulated signal. h The high-frequency signal is used to obtain the position estimation error signal f(Δθ).

[0056] Example 2

[0057] Improved quasi-proportional resonant controller in w h The Bode diagram at Figure 4 As shown, the gain characteristics at the resonant frequency are used to accurately lock the high-frequency response signal, and the transfer function is

[0058] The position estimation error signal is used as the input of the position observation loop PI regulator to estimate the rotor angular velocity and then integrate it to get the estimated rotor position. ref The difference between the estimated rotor angular velocity and the speed loop PI regulator is used as the input, and its output is the given q-axis current i qref Until the position estimation error signal f(Δθ)=0.

[0059] Example 3

[0060] The rotor position detection method based on the pulse high-frequency voltage injection method of the present invention is specifically implemented according to the following steps:

[0061] Step 1: Inject a high-frequency voltage signal into the d-axis to establish a mathematical model of the permanent magnet synchronous linear motor under high-frequency excitation, and then obtain the relationship between the high-frequency voltage signal and the high-frequency current signal in the estimated coordinate system, and finally obtain the current response equation in the rotating coordinate system, which is:

[0062] In the two-phase rotating coordinate system, the voltage equation of the permanent magnet linear synchronous motor can be expressed as formula (1):

[0063]

[0064] Among them, u d and u q It is expressed as the stator voltage d-axis component and the stator voltage q-axis component; i d and i q Indicates the d-axis component of the stator current and the q-axis component of the stator current; L d and L q represents d-axis inductance and q-axis inductance; R represents stator resistance; ψ f represents the permanent magnet flux; w e represents the rotor electrical angular velocity;

[0065] Under high-frequency sinusoidal excitation, the resistor voltage drop Ri d and Ri q The cross-coupling term ω at low speed can be ignored. e L d i d and ω e L q i q can be ignored, and the simplified high-frequency mathematical model of the permanent magnet synchronous linear motor is shown in the following formula (2):

[0066]

[0067] In the formula, u dh 、u qh is the d-axis high frequency response voltage and the q-axis high frequency response voltage; i dh 、i qh is the d-axis high frequency response current and the q-axis high frequency response current; L dh , L qh are the d-axis high frequency response inductor and the q-axis high frequency response inductor.

[0068] According to the high-frequency mathematical model (Formula 2), the current response equation in the actual coordinate system can be obtained, as shown in Formula (3):

[0069]

[0070] The real coordinate system of d and q axes The voltage transformation relationship and current transformation relationship in the axis estimation coordinate system are shown in equations (4) and (5);

[0071]

[0072] In the formula, express Axis high frequency response voltage, Axis high frequency response voltage;

[0073] express Axis high frequency response current, express Axis high frequency response current;

[0074] Among them, the true value of the electrical angle θ and the estimated value The difference is Δθ, as shown in formula (6):

[0075]

[0076] d, q axis real coordinate system to The axis estimation coordinate system transformation matrix is ​​shown in formula (7):

[0077]

[0078] According to formulas (3), (4), (5), and (7), the relationship between the high-frequency voltage signal and the high-frequency current signal in the estimated coordinate system is obtained, as shown in formula (8):

[0079]

[0080] Let the mean inductance be formula (9):

[0081] L=(L dh +L qh ) / 2 (9);

[0082] The differential inductance is (10)

[0083] ΔL=(L dj -L qh ) / 2 (10);

[0084] Substituting the mean inductance and differential inductance expressed by equations (9) and (10) into equation (8), we can obtain the relationship between the high-frequency voltage signal and the high-frequency current signal in the estimated coordinate system, as shown in equation (11):

[0085]

[0086] exist The shaft injects a high-frequency sinusoidal voltage, as shown in equation (12):

[0087]

[0088] Among them, u h is the amplitude of the injected high-frequency sinusoidal voltage, ω h is the angular frequency of the injected high-frequency sinusoidal voltage, and t is the time;

[0089] Substituting equation (12) into equation (11), we get the current response equation in the rotating coordinate system, as shown in equation (13):

[0090]

[0091] The high-frequency current response contains the rotor position error signal Δθ; in order to extract the position information, the current response needs to be demodulated.

[0092] Step 2: Using the improved quasi-proportional resonant controller Shaft current Extract high frequency response current from Then with the sinusoidal signal sin(w h t) multiplied by modulation, and another improved quasi-proportional resonant controller was used to extract a frequency of 2w h After a subtraction operation, the rotor position error signal f(Δθ) is obtained, as shown in equation (14) and equation (15).

[0093]

[0094] In the formula, K pr is the corresponding resonance parameter at high frequency, ω h is the angular frequency of the injected signal, ω c is the bandwidth cut-off frequency, The subscript ω h , 2ω h The frequencies of the signals extracted by the improved quasi-proportional resonant controller correspond to each other, and the gain characteristic of the improved quasi-proportional resonant controller at the resonant frequency can accurately lock the high-frequency response signal.

[0095] Step 3: Use the rotor position error signal f(Δθ) as the input of the position observer proportional integrator to obtain the estimated rotor angular velocity, and integrate the estimated rotor angular velocity to obtain the estimated rotor position.

[0096] Example 4

[0097] Furthermore, the rotor position error signal f(Δθ) is used as the input of the proportional integrator of the position observer to obtain the estimated rotor angular velocity, as shown in equation (16):

[0098]

[0099] The estimated rotor angular velocity is integrated to obtain the estimated rotor position, as shown in equation (17):

[0100] θ e =∫ω e (17)

[0101] Example 5

[0102] The block diagram of the improved quasi-proportional resonant controller is as follows: Figure 5 As shown, input Output is high frequency response current like Figure 4 As shown, the system has a large gain at the resonant frequency point, and the gain at other points is attenuated exponentially, which can well screen out high-frequency response signals and has no phase deviation, thus proving that the system has better response capabilities.

[0103] Example 6

[0104] Figure 6 is the waveform diagram of the actual rotor position and the estimated rotor position of the present invention, Figure 7 is the rotor position error waveform, Figure 8 is the actual speed and estimated speed waveform, from Figure 6 and Figure 7 It can be seen that the estimated rotor position quickly catches up with the actual rotor position. At the beginning, the rotor position error is large, about 0.1rad, and after a short time, the error quickly approaches 0. Figure 8 It can be seen that the estimated speed also quickly catches up with the actual speed. Based on this, it can be seen that the rotor position and speed estimation method of the present invention can quickly eliminate the rotor position error, improve the control accuracy of the system, and the parameter setting process is also simple.

[0105] To summarize, the present invention is based on a rotor position detection method based on pulsed high-frequency voltage injection, designs a permanent magnet synchronous linear motor drive system based on rotor position detection based on pulsed high-frequency voltage injection, and uses an improved quasi-proportional resonant controller to build a rotor position observer to directly demodulate the position error, thereby eliminating the traditional high-frequency injection method's dependence on filters, solving phase delay, parameter sensitivity and hardware cost problems, and improving the system's dynamic response and robustness.

Claims

1. A rotor position detection method based on a pulse high-frequency voltage injection method, characterized in that: Follow the steps below to implement it: Step 1, inject a high-frequency voltage signal into the d-axis, establish a mathematical model of the permanent magnet synchronous linear motor under high-frequency excitation, and then obtain the relationship between the high-frequency voltage signal and the high-frequency current signal in the estimated coordinate system, and finally obtain the current response equation in the rotating coordinate system; Step 2: Using the improved quasi-proportional resonant controller Shaft current Extract high frequency response current from Then with the sinusoidal signal sin(w h t) multiplied by modulation, and another improved quasi-proportional resonant controller was used to extract a frequency of 2w h The signal component is then subtracted once to obtain the rotor position error signal f(Δθ); Step 3: Use the rotor position error signal f(Δθ) as the input of the position observer proportional integrator to obtain the estimated rotor angular velocity, and integrate the estimated rotor angular velocity to obtain the estimated rotor position.

2. The rotor position detection method based on the pulse high-frequency voltage injection method according to claim 1, characterized in that: In the step 1, specifically: Under high-frequency sinusoidal excitation, the high-frequency mathematical model of the permanent magnet synchronous linear motor is obtained as shown in the following formula (2): In the formula, u dh 、u qh is the d-axis high frequency response voltage and the q-axis high frequency response voltage; i dh 、i qh is the d-axis high frequency response current and the q-axis high frequency response current; L dh , L qh are the d-axis high frequency response inductance and the q-axis high frequency response inductance; According to the high-frequency mathematical model, the current response equation in the actual coordinate system can be obtained, as shown in formula (3): The real coordinate system of d and q axes The voltage transformation relationship and current transformation relationship in the axis estimation coordinate system are shown in equations (4) and (5); In the formula, express Axis high frequency response voltage, Axis high frequency response voltage; express Axis high frequency response current, express The axis high frequency response current; Δθ is the difference between the actual value θ and the estimated value of the electrical angle difference; d, q axis real coordinate system to The axis estimation coordinate system transformation matrix is ​​shown in formula (7): According to formulas (3), (4), (5), and (7), the relationship between the high-frequency voltage signal and the high-frequency current signal in the estimated coordinate system is obtained, as shown in formula (8):

3. The rotor position detection method based on the pulse high-frequency voltage injection method according to claim 2, characterized in that: Let the mean inductance be formula (9): L=(L dh +L qh ) / 2 (9); The differential inductance is (10) ΔL=(L dh -L qh ) / 2 (10); Substituting the mean inductance and differential inductance expressed by equations (9) and (10) into equation (8), we can obtain the relationship between the high-frequency voltage signal and the high-frequency current signal in the estimated coordinate system, as shown in equation (11):

4. The rotor position detection method based on the pulse high-frequency voltage injection method according to claim 3 is characterized in that: exist The shaft injects a high-frequency sinusoidal voltage, as shown in equation (12): Among them, u h is the amplitude of the injected high-frequency sinusoidal voltage, ω h is the angular frequency of the injected high-frequency sinusoidal voltage, and t is the time; Substituting equation (12) into equation (11), we get the current response equation in the rotating coordinate system, as shown in equation (13):

5. The rotor position detection method based on the pulse high-frequency voltage injection method according to claim 4, characterized in that: In the step 2, As shown in formula (14): In the formula, K pr is the corresponding resonance parameter at high frequency, ω h is the angular frequency of the injected signal, ω c is the bandwidth cut-off frequency.

6. The rotor position detection method based on the pulse high-frequency voltage injection method according to claim 5, characterized in that: In step 2, f(Δθ) is as shown in formula (15): In the formula, The subscript ω h , 2ω j They respectively correspond to the frequencies of the extracted signals required to improve the quasi-proportional resonant controller.

7. The rotor position detection method based on the pulse high-frequency voltage injection method according to claim 6, characterized in that: In step 3, the rotor position error signal f(Δθ) is used as the input of the position observer proportional integrator to obtain the estimated rotor angular velocity, as shown in formula (16):

8. The rotor position detection method based on the pulse high-frequency voltage injection method according to claim 7, characterized in that: In step 3, the estimated rotor angular velocity is integrated to obtain the estimated rotor position, as shown in formula (17): i e =∫ω e (17)。

Images