A built-in permanent magnet linear synchronous motor position sensorless control method

Abstract

The application provides a built-in permanent magnet linear synchronous motor position sensorless control method and relates to the technical field of motor control. The application estimates the direct-axis and quadrature-axis inductance by using input-output feedback linearization adaptive control, and uses the inductance for the calculation of the given value of the direct-axis and quadrature-axis current in the maximum thrust current ratio control and the flux linkage estimation. In addition, the integral term in the flux linkage estimation operation makes the orthogonal characteristics of the flux linkage and counter electromotive force disappear, resulting in the motor position estimation error in the sensorless control. The direct current offset error compensation method is used to construct the flux linkage estimator based on the direct current offset compensation to estimate the motor flux linkage, position and speed. The maximum thrust current ratio method can output smaller current under the same thrust condition, reduce the loss of the permanent magnet linear synchronous motor, improve the efficiency level, and improve the dynamic response performance of the permanent magnet synchronous linear motor control system, improve the motor position tracking accuracy and improve the efficiency.

Description

TECHNICAL FIELD

[0001] The application relates to the technical field of motor control, in particular to a built-in permanent magnet linear synchronous motor position sensorless control method. BACKGROUND

[0002] Compared with the rotating synchronous motor, the permanent magnet linear synchronous motor (PMLSM) has no problems such as ball screw and mechanical coupling, has small mechanical friction loss and high thrust density. In order to realize high-precision vector control of the PMLSM, a mechanical sensor is used to obtain speed and position information. However, the mechanical sensor is easily affected by factors such as installation conditions, environmental temperature changes and special working conditions, the reliability of the whole system is reduced, and the cost of the motor control system is also increased. The development of motor control opens up a new route. Therefore, the permanent magnet linear motor position sensorless control technology is an important way to realize high-speed, high-precision and high-thrust control.

[0003] Due to the urgent need of high-speed and high-precision machining technology for the manufacturing of key parts of major equipment such as aerospace, new energy and microelectronics, the traditional permanent magnet linear motor cannot have the contradiction between the high thrust density of the core structure and the low thrust fluctuation of the coreless structure. In order to balance the high thrust density and the low thrust fluctuation, many researchers have proposed various new topological structures of permanent magnet linear motors, and part of them is the built-in structure of permanent magnet. For this kind of motor, due to the asymmetry of the magnetic circuit, the direct-axis and quadrature-axis inductances are different, and the reluctance torque is generated. If the traditional i d =0 control is used, the part of the reluctance torque cannot be used, resulting in the loss of motor efficiency. For the traditional rotating motor, the maximum torque current ratio can solve the above problem, but for the linear motor, there are few studies. How to construct a maximum thrust torque ratio control model suitable for the linear motor is one of the problems in the field. When realizing the position sensorless control of the built-in permanent magnet linear synchronous motor, the permanent magnet flux linkage can be calculated by using the voltage model or the current model. The current model needs the rotor position information when calculating the permanent magnet vector, therefore, the flux linkage method generally uses the voltage model or the voltage and current model to calculate the change of the synthesized flux linkage vector, and then integrates it to obtain the estimated value of the synthesized flux linkage vector, and then subtracts the stator flux linkage to obtain the estimated value of the rotor flux linkage. Generally, the stator flux linkage and the back electromotive force are in a orthogonal relationship. Since the integral term is contained in the flux linkage estimation operation, the generation of the direct current offset makes the orthogonal characteristics of the flux linkage and the back electromotive force disappear, that is, the angle between the two vectors is no longer a right angle, which leads to the estimation error of the flux linkage, which is particularly obvious at low speed. In addition, whether in the maximum thrust current ratio control or in the flux linkage estimation method, the estimated value of the inductance is needed, therefore, how to obtain the accurate inductance estimation value and how to compensate the direct current offset error of the integrator to obtain the accurate flux linkage value are the second problems to be solved in the field. SUMMARY

[0004] In view of the problems of the prior art, the application provides a built-in permanent magnet linear synchronous motor position sensorless control method.

[0005] A built-in permanent magnet linear synchronous motor position sensorless control method comprises the following steps:

[0006] Step 1: obtaining three-phase currents i A , i B , i C and three-phase voltages u A , u B , u C of the stator winding;

[0007] Step 2: performing Clark transformation on the three-phase currents and the three-phase voltages to obtain the components i α , i β and u α , u β of the voltages and currents in the two-phase static coordinate system;

[0008] Step 3: constructing a DC offset compensation-based α, β two-phase static coordinate system flux estimator to estimate the fluxes ψ α , ψ β of the α, β axes and convert the fluxes ψ α , ψ β of the α, β axes into d, q axis fluxes by coordinate conversion;

[0009] Step 3.1: estimating the DC offset error amount Δ as follows:

[0010] △=(ψ α ·ε α +ψ β ·ε β ) / |ψ s |

[0011] In the formula, the back EMF components of the α, β axes are ε α =u α -r s ·i α , ε β =u β -r s ·i β , r s is the primary armature winding resistance, and ψ s is the linear motor primary flux;

[0012] Step 3.2: adjusting the flux compensation amount ΔΨ by PI control as follows:

[0013]

[0014] where k p , k i are the proportional, integral coefficients in the proportional-integral controller PI;

[0015] Step 3.3: Estimate the flux linkage ψ α , ψ β by the flux linkage estimator

[0016]

[0017] Step 3.4: Determine the primary flux linkage phase θ s by the inverse tangent function

[0018]

[0019] Step 3.5: Calculate the estimated values of the d, q axis flux linkage and the position angle θ

[0020]

[0021] where θ s is the phase, ψ f is the permanent magnet flux linkage, |ψ s | is the primary flux linkage amplitude, ψ α and ψ β are the estimated values of the α, β axis output flux linkage, is the estimated value of the d, q axis inductance;

[0022] Step 4: Construct an adaptive input-output feedback linearization inductance estimator to estimate the inductance L d and L q , design the input-output feedback linearization control inputs γ1, γ2 and the estimated value

[0023] Select the flux linkage ψ d , ψ q and the electrical angular velocity ω re of the d, q axis as the state variables, i.e.: X = [x1 x2 x3] T = [ψ d ψ q ω re ] T Similarly, select the flux linkage ψ d , ψ q as the output variables, i.e.: Y = [y1 y2] T = [ψ d ψ q ] T , and select the input variable as U = [u1 u2] T = [u d u q ]T The input-output affine nonlinear system of the built-in permanent magnet linear synchronous motor is:

[0024]

[0025] wherein,

[0026] wherein, τ is the pole pitch; M is the mass of the moving part of the motor; F L is the load; B is the viscous friction coefficient; i d , i q is the d, q-axis current;

[0027] The designed control input is U:

[0028]

[0029] wherein,

[0030] The designed input-output feedback linearization control input γ1, γ2 is:

[0031]

[0032]

[0033] The designed flux error variables e1, e2 are the difference between the flux estimation value and the given value of the flux : The derivative is:

[0034]

[0035] Substitute the control input γ1, γ2 into the flux error to obtain:

[0036]

[0037] After the input-output feedback linearization method, the original nonlinear system is converted into a linear system, and by designing the control law and selecting the normal numbers α1, α2, the flux error of the system converges to zero;

[0038] The specific process of the adaptive input-output feedback linearization inductance estimator for estimating is as follows:

[0039] The dq-axis inductance error L1, L2 is selected as and The flux error state is

[0040]

[0041] wherein,

[0042] The Lyapunov function is selected as:

[0043]

[0044] In the formula, Γ = diag [β1, β2] is a diagonal matrix, and β1 and β2 are positive gains.

[0045] The adaptive law is designed as:

[0046]

[0047] If α1 and α2 are selected as normal numbers, and β1 and β2 are coefficients, so that

[0048] Step 5: Construct the maximum thrust current ratio controller, and calculate the given value i * d , i * q ;

[0049] When the current amplitude is constant, the thrust reaches the maximum, and the thrust F e is expressed as:

[0050]

[0051] In the formula, τ is the pole pitch, n p is the pole pair number, and according to the components of the stator current in the dq axis, the ratio f (δ) of the motor thrust to the current and its derivative are written as:

[0052]

[0053]

[0054] δ is the included angle between the primary winding flux linkage and the secondary permanent magnet flux linkage;

[0055] The maximum thrust current ratio controller calculates the given value i * d , i * q as follows:

[0056]

[0057]

[0058] Step 6: Calculate the given value Ψ * d , Ψ * q :

[0059]

[0060] Step 7: calculate the dq axis voltage given value V * d , V * q :

[0061]

[0062] In the formula, is the d, q axis current given value; is the electrical angular velocity estimated value;

[0063] Step 8: convert the dq axis voltage into ABC three-phase voltage signal u by coordinate conversion method A , u B , u C Drive the inverter to work; complete the built-in permanent magnet linear synchronous motor position sensorless control method.

[0064] Beneficial effects:

[0065] The application provides a built-in permanent magnet linear synchronous motor position sensorless control method, which applies input-output feedback linearization adaptive control method and maximum thrust current ratio method to built-in permanent magnet linear synchronous motor position sensorless control, estimates inter-axis inductance by using input-output feedback linearization adaptive control, and uses the inductance for inter-axis current given value calculation and flux estimation in maximum thrust current ratio control; the integral term in flux estimation operation makes the orthogonal characteristics of flux and counter electromotive force disappear, resulting in flux estimation error, and a direct current offset error compensation method is used to construct a flux estimator based on direct current offset compensation to estimate motor flux, position and speed. Therefore, by combining the characteristics of built-in permanent magnet linear synchronous motor, the above method is used for designing the position sensorless control system, the dynamic response performance of the permanent magnet synchronous linear motor control system is improved, and the motor position has strong robustness. BRIEF DESCRIPTION OF DRAWINGS

[0066] Figure 1 It is a kind of built-in permanent magnet linear synchronous motor position sensorless control principle block diagram of the application;

[0067] Figure 2 It is a kind of built-in permanent magnet linear synchronous motor position sensorless control flow chart of the application;

[0068] Figure 3 It is a kind of flux estimator structure diagram based on direct current offset compensation method of the application;

[0069] Figure 4A speed simulation result diagram of the built-in permanent magnet linear synchronous motor according to the present application;

[0070] Figure 5 A position simulation result diagram of the built-in permanent magnet linear synchronous motor according to the present application;

[0071] Figure 6 A d-axis estimated inductance simulation result diagram of the built-in permanent magnet linear synchronous motor according to the present application;

[0072] Figure 7 A q-axis estimated inductance simulation result diagram of the built-in permanent magnet linear synchronous motor according to the present application;

[0073] Figure 8 A d-axis and q-axis position error simulation result diagram of the built-in permanent magnet linear synchronous motor according to the present application;

[0074] Figure 9 A thrust simulation result diagram of the built-in permanent magnet linear synchronous motor according to the present application;

[0075] Figure 10 A q-axis current simulation result diagram of the built-in permanent magnet linear synchronous motor according to the present application. DETAILED DESCRIPTION

[0076] The specific embodiments of the present application will be described in detail below with reference to the accompanying drawings.

[0077] A built-in permanent magnet linear synchronous motor position sensorless control method, as shown in Figure 1 、 Figure 2 , is a built-in permanent magnet linear synchronous motor position sensorless control principle block diagram; Figure 1 Figure 1 In the figure, PI represents a proportional integral regulator, and SVPWM represents a space vector pulse width modulation; Figure 2 is a built-in permanent magnet linear synchronous motor position sensorless control flow chart. The present application includes the following steps:

[0078] Step 1: Obtain the three-phase currents i A , i B , i C and three-phase voltages u A , u B , u C of the stator winding;

[0079] Step 2: Perform Clark transformation on the three-phase currents and three-phase voltages to obtain the voltage and current components i α , i β and u α , u β in the two-phase stationary coordinate system;

[0080] ​Step 3: Construct the DC offset compensation based on the α, β two-phase static coordinate system flux estimator, as shown in Figure 3 Figure 1, to estimate the α, β axis flux ψ α , ψ β , and use coordinate conversion to convert the α, β axis flux ψ α , ψ β into d, q axis flux

[0081] Step 3.1: Estimate the DC offset error amount Δ as:

[0082] △=(ψ α ·ε α +ψ β ·ε β ) / |ψ s |

[0083] In the formula, the back EMF component of the α, β axis is ε α =u α -r s ·i α , ε β =u β -r s ·i β , r s is the primary armature winding resistance, and ψ s is the primary flux of the linear motor;

[0084] Step 3.2: Adjust the flux compensation amount ΔΨ by PI control as:

[0085]

[0086] In the formula, k p , k i are the proportional and integral coefficients in the proportional-integral controller PI;

[0087] Step 3.3: Estimate the flux ψ α , ψ β by the flux estimator:

[0088]

[0089] Step 3.4: Determine the primary flux phase θ s using the arctangent function:

[0090]

[0091] Step 3.5: Calculate the estimated value of the d, q axis flux and the position angle

[0092]

[0093] where θ s is the phase, ψ f is the permanent magnet flux linkage, |ψ s | is the primary flux linkage amplitude, ψ α and ψ β are the estimated values of the α, β axis output flux linkage, are the estimated values of the d, q axis inductances;

[0094] Step 4: Construct an adaptive input-output feedback linearization inductance estimator to estimate the inductance L d and L q , design input-output feedback linearization control inputs γ1, γ2 and estimate

[0095] Select the flux linkage ψ d , ψ q and the electrical angular velocity ω re as the state variables, that is: X = [x1 x2 x3] T = [ψ d ψ q ω re ] T Similarly, select the flux linkage ψ d , ψ q as the output variables, that is: Y = [y1 y2] T = [ψ d ψ q ] T , select the input variable as U = [u1 u2] T = [u d u q ] T ; the input-output affine nonlinear system of the interior permanent magnet linear synchronous motor is:

[0096]

[0097] where,

[0098] where τ is the pole pitch; M is the mass of the motor moving part; F L is the load; B is the viscous friction coefficient; i d , i q are the d, q axis currents;

[0099] Design the control input as U:

[0100]

[0101] where,

[0102] The design input and output feedback linearization control input γ1, γ2 is:

[0103]

[0104]

[0105] The design flux error variable e1, e2 is the flux estimation value and the difference between the flux given value The derivative is:

[0106]

[0107] Substitute the control input γ1, γ2 into the flux error to obtain:

[0108]

[0109] After the input and output feedback linearization method, the original nonlinear system is converted into a linear system, and by designing the control law and selecting the normal numbers α1, α2, the system flux error converges to zero;

[0110] The specific process of the adaptive input and output feedback linearization inductance estimator is:

[0111] The dq-axis inductance error L1, L2 is selected as and The flux error state is

[0112]

[0113] In the formula,

[0114] The Lyapunov function is selected as:

[0115]

[0116] In the formula, Γ=diag[β1, β2] is a diagonal matrix, and β1 and β2 are positive gains.

[0117] The adaptive law is designed as:

[0118]

[0119] If α1 and α2 are selected as normal numbers, and β1 and β2 are coefficients, so that

[0120] Step 5: Construct a maximum thrust current ratio controller to calculate the dq-axis current value given value i * d , i​​* q ;

[0121] The thrust reaches a maximum when the current amplitude is constant, the thrust F e is expressed as:

[0122]

[0123] where τ is the pole pitch, n p is the number of pole pairs, the ratio f(δ) of the motor thrust to the current and its derivative are written as:

[0124]

[0125]

[0126] δ is the angle between the primary winding flux and the secondary permanent magnet flux;

[0127] The maximum thrust current ratio controller calculates the current value given value i * d , i * q as follows:

[0128]

[0129]

[0130] Step 6: Calculate the given value of the flux Ψ * d , Ψ * q :

[0131]

[0132] Step 7: Calculate the dq-axis voltage given value V * d , V * q :

[0133]

[0134] where, is the current given value of the d, q axes; is the electrical angular velocity estimate;

[0135] Step 8: Convert the dq-axis voltage to the ABC three-phase voltage signal u A , u B , u CThe drive inverter works; complete built-in permanent magnet linear synchronous motor position sensorless control method.

[0136] In this embodiment, the maximum thrust current ratio based on the built-in permanent magnet linear synchronous motor position sensorless control simulation model is built in MATLAB / SIMULINK, and the simulation parameters are as follows: the stator winding resistance is 2.4Ω, the rotor mass is 1.1kg, the pole pitch is 24mm, the PI control coefficient k p and k i are 0.041 and 0.11 respectively. Figure 4 For the estimation of speed and given speed simulation results, it can be seen from the figure that the estimated speed and the actual speed are basically coincided, Figure 5 、 Figure 6 、 Figure 7 and Figure 8 are the motor position simulation results, position error simulation results, inductance estimation simulation results respectively. When the direct current offset compensation method and the input-output feedback linearization inductance estimation method are applied, the direct current bias is eliminated, the given position and the actual position curve are basically coincided, and the accuracy of the proposed method is further verified. And after applying the load at 0.3s, the position can still quickly follow, which verifies the good dynamic characteristics of the proposed method. Figure 9 For the thrust simulation waveform, Figure 10 is the q-axis current simulation result when i d =0 respectively using the maximum thrust current ratio and i When the thrust is 7N, the peak-to-peak value of the output current of the two control methods is not much different, and the maximum thrust current ratio control method and id=0 control method are 1A and 1.3A respectively, and when the thrust increases to 28N, the current peak-to-peak value is 12.5A and 19A respectively, so when the thrust is larger, the maximum thrust current ratio control method is more effective, and the efficiency is higher. The simulation results prove that the maximum thrust current ratio based on the built-in permanent magnet linear synchronous motor position sensorless control method can estimate the inductance value in real time through the input-output feedback linearization method, and then estimate the motor position and flux through the direct current offset flux estimator, and real-time compensate the motor position estimation error caused by inductance and direct current offset. The position error changes little when the load changes, and the dynamic tracking performance is good.

Claims

1. A position sensorless control method for an interior permanent magnet linear synchronous motor, characterized by, The method comprises the following steps: Step 1: Obtain three-phase currents i of the stator winding A B C and three-phase voltages u A B C ;​​​​ Step 2: Perform Clark transformation on the three-phase current and three-phase voltage to obtain the voltage and current components i in the two-phase stationary coordinate system. α i β and u α u β ; Step 3: Construct the α, β two-phase static coordinate system flux estimator based on DC offset compensation, estimate the flux ψ of α, β axis α , β , and adopt coordinate conversion to convert the flux ψ of α, β axis α , β into d, q axis flux , ; The step 3 specifically comprises: Step 3.1: estimating the direct current offset error amount Δ as: ； In the formula, the counter electromotive force component of the α, β axis is ε α = u α - r s · i α , ε β = u β - r s · i β , r s is the primary armature winding resistance, ψ s is the primary flux linkage of the linear motor; Step 3.2: adjusting the flux linkage compensation amount ΔΨ by PI control as: ； In the formula, k p , k i is the proportional, integral coefficient in the proportional-integral controller PI; Step 3.3: Estimate flux linkage ψ by flux linkage estimator α , ψ β : ， ； Step 3.4: Determine the primary flux linkage phase θ using the arctangent function s : ； Step 3.5: Calculate the estimated values of the d, q-axis fluxes and the position angle : ， ， ； where θ s is the phase, ψ f is the permanent magnet flux linkage, |ψ s | is the primary flux linkage amplitude, ψ α and ψ β are the estimated values of the α, β axis output flux linkage, , are the d, q axis inductance estimates; Step 4: Constructing adaptive input-output feedback linearization inductance estimator to estimate inductance L d With L q , design input-output feedback linearization control input γ1, γ2 and estimate 、 ; The step 4 specifically is: The flux linkage ψ d , ψ q and the electrical angular velocity ω re are selected as state variables, that is, X = [x1 x2 x3] T = [ψ d ψ q ω re ] T , and the flux linkage ψ d , ψ q are also selected as output variables, that is, Y = [y1 y2] T = [ψ d ψ q ] T , and the input variable is selected as U = [u1 u2] T = [u d u q ] T ; the input and output affine nonlinear system of the built-in permanent magnet linear synchronous motor is: ； In the formulae, ; ; where τ is the pole pitch; M is the mass of the moving parts of the motor; F L is the load; B is the viscous friction coefficient; i d , q is the d, q-axis current; Designing the control input as U: ； In the formulae, ; Designing the input-output feedback linearization control input γ1, γ2 as: ； ； The flux linkage error variables e1, e2 are designed as the difference between the flux linkage estimation values , and the flux linkage given values , The derivative is , . ； Substituting the control input γ1, γ2 into the flux linkage error to obtain: ； After the input-output feedback linearization method, the original nonlinear system is converted into a linearized system, and the system flux linkage error is converged to zero by designing the control law and selecting the normal numbers α1, α2; The adaptive input / output feedback linearized inductor estimator estimates , The specific process is as follows: The dq-axis inductance errors L1, L2 are selected as With , the flux linkage error state is ； In the formulae, , , , ; The Lyapunov function is selected as: ； In the formula, Γ=diag[β1, β2] is a diagonal matrix, and β1 and β2 are positive gains; The adaptive law is designed as: ； If a1 and a2 are selected as normal numbers, and b1 and b2 are coefficients, such that ; Step 5: Construct the maximum thrust current ratio controller, calculate the dq axis current value given value i * d , i * q ; Step 6: Calculate the given value of flux linkage, Ψ * d Ψ * q : ； Step 7: Calculate dq-axis voltage command V * d , V * q : ; where , is the d, q axis current given value; is the electrical angular velocity estimate; Step 8: Convert the dq-axis voltage to ABC three-phase voltage signal u by coordinate conversion method A B C Drive the inverter to work; complete the position sensorless control method of the built-in permanent magnet linear synchronous motor.​​ 2. The position sensorless control method of an IPMSM according to claim 1, wherein The step 5 is specifically: when the current amplitude is constant, the thrust reaches the maximum, the thrust F e is represented as: ； where τ is the pole pitch, n p The ratio f(δ) of the motor thrust to the current and its derivative ḟ(δ) are written as a function of the stator current components in the dq axes: ， ； δ is the included angle between the primary winding flux linkage and the secondary permanent magnet flux linkage; The maximum thrust current ratio controller calculates a current value given value i * d , i * q As follows: ， 。

Images