A deadbeat control method for linear Vernier permanent magnet motor

Abstract

The application relates to a kind of linear vernier permanent magnet motor deadbeat control methods, on the basis of traditional DTC control, by introducing speed backstepping controller to replace traditional speed loop PI regulator, using deadbeat controller to replace current loop PI regulator, using improved double-sampling double-update SVPWM to replace traditional SVPWM modulation module.According to multiple simulation, the application can effectively reduce the overshoot of the system and effectively reduce the thrust ripple of the system on the basis of maintaining traditional DTC control, and improve the dynamic performance of the system.

Description

TECHNICAL FIELD

[0001] The present application relates to linear vernier permanent magnet motor control technology, in particular to a kind of linear vernier permanent magnet motor deadbeat control method based on speed backstepping controller, suitable for long-stroke linear traction equipment. BACKGROUND

[0002] The rapid development of national economy and the rapid advancement of urbanization process have improved people's living standards, but the imbalance and incoordination in social and economic development are still prominent. The influx of a large number of population into cities makes the urban population more and more, the number of cities continues to rise, and the urban traffic congestion problem is becoming more and more prominent. Urban rail transit has become the main way to solve urban traffic problems with the advantages of large capacity, less pollution, low energy consumption and fast and on time.

[0003] As a core component of urban rail transit, the traction motor for vehicle is the key technology to ensure the safe and reliable operation of locomotive, and its research has also attracted extensive attention from scholars in the relevant field. In terms of technology, the traction motor for vehicle can mainly adopt two technical approaches of rotary motor and linear motor. Rotary motor drive needs to rely on the adhesion between wheels and rails, which limits the starting, acceleration, deceleration, climbing and braking performance to some extent, and the transmission loss and vehicle noise are also large, which has certain requirements for the height of the wheel, and it is difficult to realize the lightweight and miniaturization of the train. The rail transit system using linear motor drive can convert rotary motion into linear motion without intermediate energy conversion device, has the advantages of high energy conversion efficiency, low noise, simple structure and strong climbing ability, and the vehicle can obtain stronger traction and braking performance and stable running. Linear vernier permanent magnet (LVPM) motor is a new type of special motor developed on the basis of linear primary permanent magnet motor, which works by using magnetic flux switching principle, and the difference lies in that the stator teeth end is not a pair of permanent magnets but several pairs of permanent magnets. This motor can generate a larger thrust at low speed by using its vernier effect, and is widely used in long-stroke fields such as rail transit. Therefore, it has important scientific, economic and social practical value to study the linear motor system suitable for linear traction.

[0004] Direct torque control (DTC) is a mature control strategy, without decoupling calculation of the controlled model, with good dynamic performance, good robustness, simple control law and other advantages. Direct torque control is to use simple hysteresis comparator to complete decoupling control by taking flux linkage and electromagnetic torque as control variables, but torque and flux linkage pulsation is large, and adjustability is poor. LVPM motor system is a multivariable, strong coupling, nonlinear complex system, and the traditional DTC control performance is poor. In the traditional speed loop, PI regulator is generally used, but the output of the inverter voltage is limited, the thrust and speed tracking of the linear vernier permanent magnet motor are often large in pulsation. When the speed has a large mutation, the controller output will be saturated, at this time the motor can only output the preset maximum thrust, which will cause the integral saturation phenomenon, the system overshoot is large, and the stable time is long. The traditional DTC controls the flux linkage and thrust by using current hysteresis control, only a single voltage vector can be used in one period, and there are problems of large flux linkage and thrust pulsation and fixed inverter switching frequency.

[0005] The goal of deadbeat direct torque control is to make the error of torque and flux linkage with the given value zero at the end of a sampling period, and it has fast response speed and is easy to realize on high-speed microprocessor chips. Domestic and foreign scholars have researched and improved the deadbeat control algorithm, some scholars have theoretically designed the optimal reference voltage vector of the motor flux and torque deadbeat control, but the complexity is increased, the calculation of state variables is large, and the physical meaning is not clear. Some scholars have researched the design and implementation of deadbeat direct torque control under the condition that the output voltage of the voltage source inverter is limited. Some scholars have proposed a robust predictive current control algorithm of permanent magnet synchronous motor based on the deadbeat control principle in order to reduce the delay and improve the current loop control performance of the servo system. Some scholars have proposed an improved deadbeat predictive control algorithm and a new type of induction motor deadbeat direct torque control method based on the predictive algorithm based on Lagrange interpolation. Most of the above scholars have researched the deadbeat direct torque control method of asynchronous motor, and the torque and stator flux control law can be theoretically derived, but the actual sampling period is limited, and it still cannot meet the requirement of realizing zero error in one period. SUMMARY

[0006] The application provides a linear vernier permanent magnet motor deadbeat control method, which solves the problems of slow speed response and large thrust pulsation in the traditional speed loop PI and current loop PI control in the linear vernier permanent magnet motor system.

[0007] The application provides a linear vernier permanent magnet motor deadbeat control method, which solves the problems of slow speed response and large thrust pulsation in the traditional speed loop PI and current loop PI control in the linear vernier permanent magnet motor system.

[0008] Step 1: Obtain the actual speed v of the permanent magnet linear motor, and compare the actual speed v with the given speed v. * The difference Δv is used as a parameter of the backstepping controller; the system disturbance of the permanent magnet linear motor is obtained. As a parameter of another backstepping controller;

[0009] Step 2: Obtain the three-phase stator current i of the permanent magnet linear motor a i b i c After coordinate transformation, the current component i is obtained. α i β Obtain the three-phase stator phase voltage U of the permanent magnet linear motor. a U b U c After coordinate transformation, the voltage component u is obtained. α and u β The reference flux linkage ψ in the two-phase stationary coordinate system is calculated based on the current and voltage components. α ψ β ;

[0010] Step 3: After transforming the current components in the coordinate system, obtain the decoupled current i. q and i d According to the decoupled current i q and i d Calculate the actual thrust F e ;

[0011] Step 4: Generate a given reference thrust F using a backstep control law. e * Calculate the given reference thrust F e * With actual thrust F e The difference ΔF e The actual thrust F e i is obtained after PI regulator q * With d-axis reference current i d * ;

[0012] Step 5: Move i q * i d * The difference between the desired flux linkage and the reference flux linkage is used as the input parameter of the deadbeat controller. The given reference voltage u is obtained through deadbeat control law calculation. d u q ;

[0013] Step 6: Set the given reference voltage u d u qAfter coordinate system transformation, the reference voltage vector component u α1 β1 ;

[0014] Step 7: input the reference voltage vector component u α1 β1 to the SVPWM adjustment module, and then send the PWM signal with fixed switching frequency to the inverter through the SVPWM modulation module to drive the motor to run.

[0015] Further, the coordinate transformation is three-phase / two-phase static coordinate transformation; and the coordinate system transformation is two-phase static / two-phase rotating coordinate system transformation.

[0016] Further, the formula for calculating the reference flux in step 2 according to the current component and the voltage component is as follows:

[0017]

[0018]

[0019] wherein ψ α , ψ β is the flux in the two-phase static coordinate system, ψ s is the reference flux, R s is the stator resistance.

[0020] Further, the formula for calculating the actual thrust F e according to the decoupled current i q and i d is as follows:

[0021]

[0022] wherein F e is the electromagnetic thrust of the motor, v is the actual speed of the motor, τ s is the stator pitch of the LVPM motor, P n is the number of pole pairs of the motor, ψ f is the permanent magnet flux, L d , L q are the direct-axis component and the quadrature-axis component of the inductance in the dq coordinate system, ω r = n p πv / τ s is the motor rotor flux electric angular velocity.

[0023] Further, the backstepping control law in step 4 is as follows:

[0024]

[0025] wherein F e * is the reference thrust; e​​v To define the speed error, e v = v - v * ; F L is the load resistance; M is the primary mass of the motor; B is the friction coefficient; k is a normal number;

[0026] Further, the given reference voltage u d , u q in step 5 is obtained by the formula of the dead-beat control rate operation

[0027]

[0028]

[0029] Wherein, T s is the sampling period, and ω r is the electric angular velocity.

[0030] Further, the SVPWM adjustment module in step 7 adopts a double-sampling double-updating strategy, that is, at the beginning and middle time of each period, the sampling of the current period is carried out, and at the same time, the PWM is updated to the instruction of the last period.

[0031] The beneficial effects of the present application are:

[0032] 1. The speed loop of the double-closed-loop motor system is changed into a speed backstepping controller, the disturbance quantity of the system is estimated by using an extended state observer, the control law is designed and verified by stability, and the reference quantity of the thrust F e is obtained, so that the anti-interference and robustness of the system are improved.

[0033] 2. The dead-beat controller is adopted to replace the conventional current loop, the voltage u d , u q of the next moment is calculated according to the current moment flux and current, the control law of the voltage is given, and at the same time, the double-sampling double-updating strategy of the SVPWM modulation module is designed, so that the real-time performance of the PWM wave is improved, the whole strategy can make the LVPM motor system have good dynamic performance and speed tracking performance, and the overshoot of the system can be reduced.

[0034] 3. The present application is also applicable to common non-primary permanent magnet type linear permanent magnet motors. BRIEF DESCRIPTION OF DRAWINGS

[0035] The features and advantages of the present application can be more clearly understood by referring to the accompanying drawings, which are schematic and should not be understood as limiting the present application, and in the drawings:

[0036] Figure 1 is the system block diagram of the present application;

[0037] Figure 2 Figure 1 is a structure diagram of a speed backstepping controller;

[0038] Figure 3 Figure 2 is a comparison diagram of three PWM update strategies;

[0039] Figure 4 Figure 3 is a voltage space vector distribution diagram;

[0040] Figure 5 Figure 4 is a dynamic performance simulation waveform diagram;

[0041] Figure 6 Figure 5 is a simulation waveform diagram when thrust suddenly changes

[0042] Figure 7 Figure 6 is a flux trajectory circle under different methods;

[0043] Figure 8 Figure 7 is a speed and thrust response curve diagram when speed suddenly changes. DETAILED DESCRIPTION

[0044] In order to make the purpose, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are some embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor fall within the scope of protection of the present application.

[0045] As shown in Figure 1 , 2 , the displacement sensor measures the motor displacement in real time and obtains the actual speed v after derivation. The difference between the actual speed v and the given speed v * is taken as one of the inputs of the speed backstepping controller. The load of the system is regarded as a disturbance, and the disturbance is obtained by the extended state observer ESO as another input of the speed backstepping controller. After the control law operation of the speed backstepping controller, the reference thrust F e * is generated. At the same time, the three-phase stator current, voltage is converted into current components i α , i β , voltage components u α , u β through three-phase / two-phase static coordinate conversion, and the flux ψ s is calculated through flux estimation. Then, the motor decoupling current i q and i d are obtained through two-phase static / two-phase rotating coordinate conversion, and the thrust F e is calculated through thrust estimation. The reference thrust F e* and actual thrust F e q * as the input of the deadbeat controller, given flux reference ψ s * s q * d * as the input of the deadbeat controller; in the deadbeat controller, according to the designed voltage control law, the voltage u is calculated q d α1 β1 In the SVPWM modulation module, the double-sampling and double-updating strategy is adopted, which can reduce the delay of sampling, generate the PWM waveform output to the inverter, control the output voltage of the inverter, and realize more stable and accurate control of the speed and thrust of the motor.

[0046] The specific steps of the embodiment are as follows:

[0047] Step S1: The permanent magnet linear motor displacement s is obtained by using a position sensor, and the actual speed v of the motor is obtained by deriving the motor displacement s. The difference Δv between the actual speed v and the given speed v * is taken as the parameter of the backstepping controller; the system disturbance of the permanent magnet linear motor is obtained by an extended state observer as the parameter of another backstepping controller;

[0048] Step S2: The three-phase stator currents i a , i b , i c of the permanent magnet linear motor are obtained by a Hall sensor, and the current components i α , i β are obtained after three-phase / two-phase static coordinate transformation; the bus voltage U dc is obtained by a voltage collector, and the three-phase stator voltages U a , U b , U c of the permanent magnet linear motor are calculated respectively according to the switching states S a , S b , S c of the inverter; the voltage components u α and u β are obtained after three-phase / two-phase static coordinate transformation, and the reference flux is calculated according to the current components and the voltage components; ​​​​​​​

[0049] The coordinate transformation formula is as follows:

[0050]

[0051]

[0052] The flux linkage calculation formula is as follows:

[0053]

[0054]

[0055] Wherein, ψ α , ψ β is the flux linkage in the two-phase static coordinate system, ψ s is the reference flux linkage, R s is the stator resistance.

[0056] Step 3: After the current component is transformed again through the two-phase static / two-phase rotating coordinate system, the decoupled currents i q and i d are obtained, and the specific formula is as follows:

[0057]

[0058] Wherein, θ r is the included angle between the d-axis and the a-axis;

[0059] The actual thrust F e is calculated according to the decoupled currents i q and i d , and the specific formula is as follows:

[0060]

[0061] Wherein, F e is the electromagnetic thrust of the motor, v is the actual speed of the motor, τ s is the stator pole pitch of the LVPM motor, P n is the number of pole pairs of the motor, ψ f is the permanent magnet flux linkage, L d , L q are the direct-axis component and the quadrature-axis component of the inductance in the dq coordinate system. ω r =n p πν / τ s is the motor rotor flux linkage angular velocity.

[0062] Step 4: The given reference thrust F is generated through the backstepping control law. The backstepping control law is as follows:

[0063]

[0064] Calculate the given reference thrust The difference ΔF e of the actual thrust F e ; According to the above formula (1), the L d of the hidden pole motor q So The actual thrust F e Through the PI regulator to obtain i q * The difference between the d-axis reference current i d * ;

[0065] Step 5: i q * , i d * , the difference between the expected flux and the reference flux is taken as the input parameter of the deadbeat controller, and the given reference voltage u d , u q is obtained by deadbeat control calculation, and the specific formula is as follows:

[0066]

[0067]

[0068] Step 6: The given reference voltage u d , u q is transformed into the reference voltage vector components u α1 , u β1 after the coordinate system transformation, and the specific formula is as follows:

[0069]

[0070] Step 7: The reference voltage vector components u α , u β are input to the SVPWM adjustment module, and a double sampling and double updating strategy is adopted, as shown in Figure 3 The present application updates the PWM to the command of the last period at the same time when sampling the current period at the beginning and middle of each period, so the total delay time of the sampling PWM updating algorithm will not exceed 0.75T s . As shown in the mode 3 part of Figure 3 Table 1 is a comparison of the delay time of the three updating modes,

[0071] PWM update mode Total latency Calculation latency PWM latency Mode 1 1.5T s ]]> [CAT s ]] 0.5T s ]] Mode 2 [TECHNICAL FIELD] s ]] 0.5T s ]] 0.5T s ]] Mode 3 0.75T s ]] 0.5T s ]] 0.25T s ]]

[0072] Table 1

[0073] Then the SVPWM modulation module sends a PWM signal with fixed switching frequency to the inverter to drive the motor to run.Figure 4 The second sector in the figure is taken as an example to analyze the PWM signal modulation and duty cycle. For a given voltage vector amplitude U m The vector output vector time is calculated as follows:

[0074]

[0075] To verify the effectiveness of the application, a simulation model is built. The motor parameters are as follows: stator winding resistance R = 1.25 Ω, pole pair number p n = 2, L d = 0.0851 H, L q = 0.0852 H, permanent magnet flux linkage Ψ f = 0.11 Wb, pole pitch τ = 0.00147 m, viscous friction coefficient B v = 0.1 N.s / m, motor primary mass M r = 35 kg, initial load thrust F l = 120 N, rated voltage U N = 270 V, rated current I N = 3.5 A, rated speed ω e = 1.5 m / s.

[0076] Figure 5 (a1)- Figure 5 (a3) is a dynamic performance simulation diagram when the speed of the application method is suddenly changed, Figure 5 (b1)- Figure 5 (b3) is a dynamic performance simulation diagram when the speed of the PI method is suddenly changed,

[0077] Taking the application method as an example, given the speed 0.2 m / s, when t = 0.1 s, the motor speed reaches and stabilizes at 0.2 m / s;

[0078] At t = 0.2 s, the given speed suddenly increases to 0.4 m / s, and the motor speed quickly follows, reaching 0.4 m / s after 0.025 s, i.e. t = 0.225 s;

[0079] At t = 0.4 s, the given speed suddenly increases to 0.8 m / s, and the motor speed quickly follows, reaching 0.8 m / s after 0.12 s, i.e. t = 0.52 s;

[0080] At t = 0.6 s, the given speed suddenly decreases to -0.2 m / s, and the motor speed follows, reaching -0.2 m / s after 0.15 s, i.e. t = 0.75 s;

[0081] At t = 0.8 s, the given speed suddenly increases from -0.2 m / s to 0.4 m / s, and the motor speed reaches 0.4 m / s after 0.12 s, i.e. t = 0.82 s.

[0082] It can be seen from the figure that the speed following performance of the method is better than that of the PI method, and the rapidity is stronger.

[0083] As Figure 6 is a thrust dynamic performance simulation diagram, when t=0.2s, the given load thrust is suddenly increased from 100N.m to 150N.m, when t=0.4s, the given load thrust is decreased from 150N.m to 100N.m, when t=0.6s, the given load thrust is decreased from 100N.m to 50N.m, and when t=0.8s, the given load thrust is suddenly increased from 50N.m to 100N.m, and the response time of the electromagnetic thrust is about 0.01s each time the thrust is suddenly changed.

[0084] When the thrust is suddenly increased, the reference stator flux linkage ψ s * is given by the MFPA (unit current maximum thrust), and the flux linkage amplitude is also increased when the thrust is increased; when t=0.6s, the given load thrust is suddenly decreased from 100N to 50N, and the reference stator flux linkage ψ s * is given by the MFPA, and the corresponding flux linkage amplitude is also decreased when the thrust is decreased. Obviously, the simulation results show that the method has good speed and thrust dynamic performance.

[0085] Figure 7 When the speed is 0.2m / s, the flux linkage trajectories of the method and the existing PI method are shown in the following figures: Figure 7 (a) the flux linkage trajectory circle of the method, and Figure 7 (b) the flux linkage trajectory circle of the PI method can be seen that due to the influence of the equivalent capacitor voltage, the flux linkage circle pulsation of the method is slightly larger than that of the PI method, but the flux linkage of the two modes follows a better circular trajectory, indicating that the method has good flux linkage control performance.

[0086] Figure 8 The conditions shown are the speed response curve under the condition of given initial speed command v*=0.8m / s, speed increase to 1m / s at 0.5s, DTC control method thrust response curve, and the thrust response curve of the method. Figure 8 (a) is the speed response curve, Figure 8 (a1) is Figure 7 (a) is an enlarged view of A, wherein the uppermost curve is the speed curve of the DTC control method, the lowermost curve is the speed curve of the method, and the straight line in the middle is the expected speed; Figure 8 (a2) is Figure 8 (a) is an enlarged view of B, wherein the uppermost curve is the speed curve of the DTC control method, the middle curve is the speed curve of the method, and the lowermost straight line is the expected speed. Figure 8(b) is the thrust response curve of the DTC control method, Figure 8 (c) is the thrust response curve of the method of the present application. Figure 8 It can be seen that, compared with the conventional DTC control, the speed step and the sudden change response of the method of the present application is slightly faster, and the overshoot is significantly reduced. The overshoot of the step response is reduced from 8% to 1%, and the overshoot of the sudden change is reduced from 2.5% to 1%. In the step response moment, the initial impact force of the thrust is significantly reduced, from 770 N.m to 500 N.m, and in the speed sudden change, the thrust impact force is reduced from 330 N.m to 240 N.m. It is shown that the method of the present application is better than the DTC control method, and the dynamic, rapidity and anti-interference of the system are improved.

[0087] Although the embodiments of the present application are described in conjunction with the drawings, various modifications and changes can be made by those skilled in the art without departing from the spirit and scope of the present application, and such modifications and changes fall within the scope defined by the appended claims.

Claims

1. A deadbeat control method for a linear vernier permanent magnet motor, characterized in that, Includes the following steps: Step 1: Obtain the actual speed v of the permanent magnet linear motor, and compare the actual speed v with the given speed v. * The difference e v As parameters for the backstepping controller; to obtain the system disturbance of the permanent magnet linear motor. As another parameter of the backstep controller; Step 2: Obtain the three-phase stator current i of the permanent magnet linear motor a i b i c After coordinate transformation, the current component i is obtained. α i β Obtain the three-phase stator phase voltage U of the permanent magnet linear motor. a U b U c After coordinate transformation, the voltage component u is obtained. α and u β The reference flux linkage ψ in the two-phase stationary coordinate system is calculated based on the current and voltage components. α ψ β ; Step 3: After transforming the current components in the coordinate system, obtain the decoupled current i. q and i d According to the decoupled current i q and i d Calculate the actual thrust F e ; Step 4: Generate a given reference thrust using a backstep control law. Calculate the given reference thrust With actual thrust F e The difference ΔF e The actual thrust F e i is obtained after PI regulator q * With d-axis reference current i d * , The backstep control law is as follows: Among them, F e * For reference thrust; e v To define the speed error, e v =vv * ; denoted as , where M is the primary mass of the motor; B is the coefficient of friction; and k is a constant. Step 5: Move i q * i d * The difference between the desired flux linkage and the reference flux linkage is used as the input parameter of the deadbeat controller. The given reference voltage u is obtained through deadbeat control law calculation. d u q ; Step 6: Set the given reference voltage u d u q After coordinate transformation, the reference voltage vector component u is obtained. α1 uβ1; Step 7: Convert the reference voltage vector component u α1 u β1 The signal is input to the SVPWM adjustment module, and then the SVPWM modulation module sends a PWM signal with a fixed switching frequency to the inverter to drive the motor.

2. The deadbeat control method for a linear vernier permanent magnet motor as described in claim 1, characterized in that, The coordinate transformation is a three-phase / two-phase stationary coordinate transformation; the coordinate system transformation is a two-phase stationary / two-phase rotating coordinate system transformation.

3. The deadbeat control method for a linear vernier permanent magnet motor as described in claim 1, characterized in that, The formula for calculating the reference flux linkage based on the current and voltage components in step 2 is as follows: Where, ψ α ψ β Let ψ be the magnetic flux linkage in a two-phase stationary coordinate system. s For reference flux linkage, R s This is the stator resistance.

4. The deadbeat control method for a linear vernier permanent magnet motor as described in claim 1, characterized in that, In step 3, based on the decoupling current i q and i d Calculate the actual thrust F e The formula is as follows: Among them, F e τ is the electromagnetic thrust of the motor, v is the actual speed of the motor, and τ is the electromagnetic thrust of the motor. s P is the stator pole pitch of the LVPM motor. n Let ψ be the number of pole pairs of the motor. f For permanent magnet chain, L d L q Let be the direct-axis and quadrature-axis components of the inductance in the dq coordinate system.

5. The deadbeat control method for a linear vernier permanent magnet motor as described in claim 1, characterized in that, In step 5, the given reference voltage u is obtained through deadbeat control rate calculation. d u q The formula is as follows: Among them, T s ω is the sampling period. r Let ψ be the electric angular velocity. f For permanent magnet chain, L d L q Let R be the direct-axis and quadrature-axis components of the inductance in the dq coordinate system. s This is the stator resistance.

6. The deadbeat control method for a linear vernier permanent magnet motor as described in claim 1, characterized in that, In step 7, the SVPWM adjustment module adopts a dual sampling and dual update strategy, that is, when sampling is performed at the beginning and middle of each cycle, the PWM is simultaneously updated to the instruction of the previous cycle.

Images