Motor speed control device and motor speed control method

Abstract

This invention provides a motor speed control device that enables the application of DB control to motor speed control. [Solution] The motor current control device of the embodiment includes a power converter having a plurality of semiconductor switching elements that converts DC power to AC power and supplies it to an AC motor, a current detection unit that detects the current supplied to the AC motor, a speed detection unit that detects the speed of the AC motor, and a control unit that controls the on / off state of the plurality of semiconductor switching elements. The control unit includes a speed control unit that generates a command current based on the deviation between the input command speed and the detected speed, and a current control unit that generates a command voltage based on the deviation between the command current and the detected current. The speed control unit calculates a command current for each control cycle that causes the speed of the AC motor to follow a target speed in a control cycle multiple times ahead from the current control cycle, by summing the command current calculated in the previous calculation and the correction current calculated in the current calculation, and outputs it to the current control unit.

Description

[Technical Field]

[0001] Embodiments of the present invention relate to a device and method for controlling the speed of a motor. [Background technology]

[0002] In the field of variable-speed AC motors, permanent magnet synchronous motors (PMSMs) are currently widely used due to their characteristics such as high efficiency and high torque density. For PMSMs, it is desirable that the control system's settling time be easily and quickly designed in applications requiring high responsiveness, such as servo systems. Deadbeat (DB) control is a control method that can achieve these requirements. DB control is a method that sets the deviation of the controlled variable to zero completely within a finite control period. In recent years, DB control has been applied to the current control of PMSMs, for example, as shown in Patent Document 1. [Prior art documents] [Non-patent literature]

[0003] [Patent Document 1] Japanese Patent Publication No. 2019-83673 [Overview of the project] [Problems that the invention aims to solve]

[0004] However, the application of DB control to speed control has not been sufficiently considered at present. Existing DB control used for current control is generally a method aimed at minimum time setting, and is achieved by simply solving the inverse system of the controlled object. This is based on the fact that the current detection signal is an analog value, and an A / D converter can be used to detect the signal with sufficiently high response and high resolution to the current fluctuation per control cycle.

[0005] On the other hand, when using a rotary encoder, the speed is calculated based on the detection timing of the pulse signal output from the encoder. Therefore, the responsiveness and accuracy of the detection are limited by the motor's rotational speed, the sensor's position resolution, and the arithmetic unit's clock frequency. As a result, the conditions under which an approach similar to DB current control can be applied to motor speed control are currently limited.

[0006] Therefore, the present invention provides a motor speed control device and a motor speed control method that enable the application of DB control to motor speed control. [Means for solving the problem]

[0007] The motor current control device of the embodiment includes a power converter having a plurality of semiconductor switching elements that converts DC power to AC power and supplies it to an AC motor, A current detection unit for detecting the current supplied to the AC motor, A speed detection unit for detecting the speed of the AC motor, The system comprises a control unit that controls the on / off switching of the plurality of semiconductor switching elements, The control unit includes a speed control unit that generates a command current based on the deviation between the input command speed and the detected speed, The system includes a current control unit that generates a command voltage based on the deviation between the command current and the detected current, The speed control unit calculates a command current for each control cycle, which is the sum of the command current calculated in the previous cycle and the correction current calculated in the current cycle, to make the AC motor's speed follow a target speed in a control cycle multiple cycles ahead of the current control cycle, and outputs this command current to the current control unit. [Brief explanation of the drawing]

[0008] [Figure 1] This is a functional block diagram showing the configuration of a motor speed control device, representing the first embodiment. [Figure 2] Functional block diagram showing the detailed configuration of the speed control unit. [Figure 3]A time chart showing the step speed response with ideal minimum time DB speed control. [Figure 4] Timing chart of step speed response by DB speed control in this embodiment [Figure 5] Figure showing simulation conditions and motor constants [Figure 6] This figure shows the speed step response when nc=4, as well as the commanded value and actual current of the q-axis current. [Figure 7] Figure 6 equivalent when nc=40 [Figure 8] This figure shows the speed step response using conventional PI control and the speed step response using this embodiment when nc=40 side by side. [Figure 9] This is a second embodiment, showing the case where nc=40 and the window function is a rectangular window and the case where it is a Hamming window. [Figure 10] This figure shows the velocity step response when the window function is a Hamming window. [Figure 11] Diagram showing the mathematical formula (Part 1) [Figure 12] Diagram illustrating the mathematical formula (part 2) [Figure 13] Diagram illustrating the mathematical formula (Part 3) [Modes for carrying out the invention]

[0009] (First Embodiment) The first embodiment will be described below with reference to Figures 1 to 8. Note that in the block diagram shown in Figure 1, the names of each functional block are shown in English. The speed control unit 1 receives the command speed ω via the subtractor 16. * The deviation e between the speed ω hat (^) calculated by the speed calculation unit 14 is input. The symbol ^ indicates that it is an estimated value. The speed control unit 1 also receives the q-axis current i output by the 3-phase → dq coordinate transformation unit 9 via the subtractor 10. q And the disturbance observer 15 outputs the estimated q-axis disturbance current i qD The acceleration component i of the q-axis current is the difference from the hat. qA This has been entered.

[0010] Based on the above input, as a result of performing speed DB control, the acceleration component i of the q-axis command current qA* is output to the adder 2. In the adder 2, the acceleration component i qA* is added to the estimated q-axis disturbance current i qD ^, and the result is output to the current control unit 3 as the q-axis command current i q* . A zero value is input to the current control unit 3 as the d-axis command current i d* . In addition, the q-axis current i q and the d-axis current i d output by the 3-phase → dq coordinate conversion unit 9, and the speed ω^ are input to the current control unit 3. Based on the above input, as a result of performing current DB control, the current control unit 3 outputs the q-axis voltage V q and the d-axis voltage V d to the dq → 3-phase coordinate conversion unit 4.

[0011] The electrical angle θ calculated by the angle calculation unit​​​​​​​Each phase output terminal of the inverter unit 6 is connected to one end of each phase winding of the PMSM 11, which is an AC motor. The PMSM 11 incorporates a rotary encoder 12, which is a speed detection unit. As the rotor of the PMSM 11 rotates, the rotary encoder 12 generates signals of phases A, B, and Z and outputs those signals to the angle calculation unit 13. The speed calculation unit 14 receives the signals of phases A and B. The rotary encoder 12 and the speed calculation unit 14 correspond to the speed detection unit.

[0014] The disturbance observer 15 receives the speed ω hat calculated by the speed calculation unit 14 and the q-axis command current i q* output from the adder 2. Based on these input signals, the disturbance observer 15 calculates and outputs the estimated q-axis disturbance current i qD hat. In the above, excluding the PSMS 11 constitutes the motor speed control device 17. Also, excluding the inverter unit 6, the current detection unit 7, and the rotary encoder 12 from the motor speed control device 17 constitutes the control unit 18.

[0015] Next, the details of the control content in each functional block will be described. First, the DB current control will be explained. <DB Current Control> The voltage equation in the dq-axis rotating coordinate system of a general PMSM is represented by Equation (1) shown in FIG. 11. Here, R is the winding resistance, Ld is the d-axis inductance, Lq is the q-axis inductance, ω is the electrical angular frequency, and Φ is the maximum magnetic flux linkage due to the permanent magnet. When Equation (1) is converted into a state equation, it becomes Equation (2) shown in the same figure as follows, and the system matrices A and B in Equation (2) are as shown in Equation (3).

[0016] For the state equation in equation (2), the input is the voltage minus the induced voltage, and the state variables are the d and q-axis currents. Discretizing the derived equation (2) with respect to time k and sampling period T yields equation (4). G and H in equation (4) are given in equation (5). For the matrix exponential function in equation (5), an approximation using a Maclaurin series is performed depending on the required discretization accuracy. The d and q-axis currents i at time (k+1) d k+1 ,i q k+1 of, Command value i d* k+1 ,i q* k+1 By substituting and solving for the d and q axis voltages, we obtain the control law (6) which derives the voltage required for current tracking in one period.

[0017] According to the above calculation rules, the d and q axis currents perfectly track the command value in one control cycle. However, since the above DB control has low robustness to errors in motor parameters, it is desirable to apply a robust control method such as oversampling deadbeat control or a control method combined with parameter estimation as needed.

[0018] <Disturbance Observer 15> When a relative transformation is performed using the dq transformation, the equation of motion for a typical PMSM is given by equation (7), where P is the number of pole pairs, J is the inertia, D is the coefficient of viscous friction, and τ L : This is the load torque. Disturbance torque τ D This is defined by equation (8) shown in Figure 12. Here, J-hat: inertia design value, Φ-hat: permanent magnet flux design value. Without considering reluctance torque, and assuming id=0, the disturbance torque τ is defined. D q-axis external disturbance i qD Converting to this, we obtain the same equation (9) shown in the same figure.

[0019] The q-axis current i, proportional to acceleration, after removing the disturbance component. qA =i q -i qDUsing this, and transforming equation (7) based on equations (8) and (9), the equation of motion is simplified to equation (10). Note that the q-axis external disturbance current i qD Since it is impossible to directly detect the disturbance current i, we derive it based on equation (11) according to the principle of disturbance observers. The disturbance current i obtained from equation (11) qD Because it contains estimation errors due to noise, passing it through a low-pass filter (LPF) allows for the estimation of the q-axis external disturbance current i. qD To use as a hat.

[0020] <Minimum Time DB Speed ​​Control> Figure 3 shows a time chart illustrating the step speed response under ideal minimum-time DB speed control. To ensure that there is no pulsation in speed between sample points and that the system tracks the command value in the shortest possible time, the speed and current are controlled to converge to steady-state values ​​over two control cycles, as shown in the figure. The control equation is derived by solving the discrete equation of motion from time k to (k+2). In the simplified equation of motion (10), the q-axis current i corresponds to the input. qA Since it changes linearly with a short discrete period, the equation of motion is discretized by the Tustin transform. Applying the Tustin transform to equation (10) yields equation (12).

[0021] Adding the equation obtained by advancing the time by one control cycle to equation (12) results in equation (13). Since the rotor acceleration is zero at time (k+2), i qA k+2 Let i = 0. qA k+1 , ω k+2 Each of these is the command value i qA* k+1 , ω * k+2 Let's assume the speed deviation is e k =ω * k+2 -ω k i qA* k+1 Solving for this, the command current in minimum-time DB speed control is given by equation (14).

[0022] <DB speed control of this embodiment> Figure 4 shows the timing chart of the step speed response by DB speed control in this embodiment. In this figure, the settling target period n c The period is set to 3. The period for speed control is set to 1 / 2 the carrier period in PWM control. The initial condition is that the command speed changes in a stepwise manner at time (k-1) and settles at time (k+2), as shown by the time vector i of the command current in equation (M1). qA* kー1 This is the case where the command current at time k inherits the command current vector commanded at time (k-1), and the command velocity ω * k+nc The current that is insufficient to set it is corrected by the correction current vector i shown in Figure 12 by equation (M2). qadJ k The command current vector i qA* kー1 The calculation is performed by adding to the given value. The specific calculation formula is shown below.

[0023] First, from time k to time (k+n c Derive the discrete equation of motion from time k to (n). Equation (12) from time k to (n c -1) By taking the sum of the equations obtained by advancing each period by one period, we obtain equation (15) shown in Figure 13. Time (k+n c ) Since the q-axis current converges to a steady value, i qA k+nc Let i = 0. qA k The command value i at time k is qA* k Let it be an element of ω. k+nc command speed ω * k+nc Let the speed deviation be e k =ω * k+nc -ω k By rearranging equation (15), we obtain equation (16).

[0024] Equation (16) is given by time n cThis represents the sum of the command currents required to set the speed. To implement DB speed control in this embodiment, the value of equation (16) should be made to match the sum of the command current inherited from time (k-1) and the correction current at time k. Therefore, the sum of the correction currents is given by equation (17).

[0025] The corrected current vector is given by equation (18) when all elements have the same value. Using the corrected current vector obtained from equation (18), the command current vector at time k is updated by equation (19). In this equation, U is a matrix that shifts the elements of the vector upward and sets the bottom element to 0, and is expressed by equation (20). The command value output to the current control unit 3 is i qA* k This is the first element.

[0026] Figure 2 shows the functional block in the speed control unit 1 that performs the calculation of equation (19). The speed deviation is e k The current i is input to multiplier 21, and the coefficient of the first term on the right-hand side of equation (16) is multiplied. The result of the multiplication by multiplier 21 is input to the subtractor input of subtractor 22. qA k The input is passed to multiplier 23, where it is multiplied by a coefficient of 1 / 2. The result of the multiplication by multiplier 23 is passed to adder 24. The result of the addition by adder 24 is passed to the subtraction input of subtractor 22.

[0027] The subtraction result from subtractor 22 is input to multiplier 25 and the coefficient {1 / (n c When multiplied by (-1), it is then input to multiplier 26, where a column vector with all elements being "1" is multiplied. The result of multiplication by multiplier 26 becomes the left side of equation (18) and is input to adder 27. The result of addition by adder 27 is input to accumulator 30 via delay unit 28 and multiplier 29. Multiplier 29 multiplies by matrix U shown in equation (20). The result of accumulation by accumulator 30 is input to adder 24.

[0028] Next, the results of a simulation of the DB speed control in this embodiment using MathWorks' MATLAB® / Simulink® are shown. The specifications of the PMSM used as the simulation conditions are shown in Figure 5. For simplicity, the PMSM was unloaded, the disturbance observer 15 was omitted, the carrier frequency was 10 kHz, and the speed control period was 50 μs. c =4,40 is set to n shown in Figure 6. c When = 4, the time it takes for the velocity step response to follow is 0.2 ms, as shown in Figure 7. c When = 40, the time it takes for the speed step response to follow is 2 ms. The follow time is defined as the time it takes for the actual speed to be within 5% of the speed step width of 100 rpm. Thus, according to this embodiment, the speed can be tracked at any control cycle.

[0029] Furthermore, Figure 8 shows the velocity step response using conventional PI (Proportional-Integral) control and the n according to this embodiment. c This figure shows the velocity step response for the case where =40 side by side. The tracking time for the former is 8ms, while the tracking time for the latter is 2ms, indicating that the velocity tracking time has been shortened compared to conventional methods.

[0030] As described above, according to this embodiment, the inverter unit 6 converts DC power to AC power and supplies it to the PMSM 11, the current detection unit 7 detects the phase currents iu, iv, and iw supplied to the PMSM 11, the speed calculation unit 14 calculates and detects the speed of the PMSM 11 from the A-phase and B-phase signals input from the rotary encoder 12, and the control unit 18 controls the on / off state of the multiple semiconductor switching elements constituting the inverter unit 6.

[0031] The control unit 18 then receives the command speed ω * Based on the deviation between the speed ω-hat detected by the speed calculation unit 14, the acceleration component i of the q-axis command current is calculated. qA* A speed control unit 1 outputs to adder 2, and a command current i is output via adder 2. q* , detected current iq Based on the deviation from, the command voltages V q and V d A current control unit 3 that generates is provided.

[0032] The speed control unit 1 calculates, for each control cycle, a command current i qA* that causes the speed ω̂ of the PMSM 11 to follow the target speed in a plurality of subsequent control cycles from the current control cycle, as the sum of the previously calculated command current and the correction current calculated this time, and outputs it to the current control unit 3. With this configuration, it becomes possible to apply DB control to the speed control of the PMSM 11, and the speed response characteristics can be improved compared to the conventionally applied PI control.

[0033] (Second Embodiment) Hereinafter, the same parts as those in the first embodiment will be denoted by the same reference numerals and the description thereof will be omitted, and the different parts will be described. In the first embodiment, all elements of the column vector of the correction current vector shown in Equation (18) are "1", which indicates that the window function is a rectangular window. The sum of all elements is (n c -1). That is, if a column vector w of elements weighted by the values of the window function is defined by Equation (21), as shown in FIG. 13, Equation (18) is represented by Equation (22).

[0034] In addition, if the window function is, for example, a Hamming window, the function is given by Equation (23). w(x)=0.54 - 0.46cos(2πx), 0 ≤ x ≤ 1 #(23) FIG. 9 shows the case where the window function is a rectangular window and the case where the window function is a Hamming window when n c = 40. FIG. 10 shows the speed step response when the window function is a Hamming window for the case of n c = 40. Thus, by applying a column vector w having (n c -1) elements weighted by the appropriately selected values of the window function, it is possible to reflect the characteristics according to the window function in the speed response characteristics.

[0035] (Other Embodiments) n cThe value is not limited to "4" or "40," but can be changed as appropriate depending on the individual design. The window functions that can be applied are not limited to rectangular windows or Hamming windows. The carrier frequency, speed control period, motor constants, etc., can also be changed as appropriate. For the current control unit, DB control is not necessarily required; PI control may also be applied.

[0036] While several embodiments of the present invention have been described, these embodiments are presented as examples only and are not intended to limit the scope of the invention. These novel embodiments can be implemented in a variety of other forms, and various omissions, substitutions, and modifications can be made without departing from the spirit of the invention. These embodiments and their variations are included in the scope and spirit of the invention, as well as in the claims of the invention and its equivalents. [Explanation of symbols]

[0037] In the drawing, 1 is the speed control unit, 3 is the current control unit, 5 is the PWM generation unit, 6 is the inverter unit, 7 is the current detection unit, 11 is the PMSM, 12 is the rotary encoder, 13 is the angle calculation unit, 14 is the speed calculation unit, 15 is the disturbance observer, and 18 is the control unit.

Claims

1. A power converter having multiple semiconductor switching elements that converts DC power to AC power and supplies it to an AC motor, A current detection unit for detecting the current supplied to the AC motor, A speed detection unit for detecting the speed of the AC motor, The system comprises a control unit that controls the on / off switching of the plurality of semiconductor switching elements, The control unit includes a speed control unit that generates a command current based on the deviation between the input command speed and the detected speed, The system includes a current control unit that generates a command voltage based on the deviation between the command current and the detected current, The speed control unit calculates a command current for each control cycle, which is the sum of the command current calculated in the previous cycle and the correction current calculated in the current cycle, to make the AC motor's speed follow a target speed in a control cycle multiple times beyond the current control cycle, and outputs this command current to the current control unit.

2. The motor speed control device according to claim 1, wherein the current control unit calculates and outputs a command voltage that makes the deviation between the command current and the detected current zero in the next control cycle from the current control cycle.

3. The speed control unit determines the command current as the vector sum of the previous command current and the correction current, The aforementioned plurality of values ​​n ｃ If so, the vector matrix of the correction current is the first to the (n) elements, each of which is weighted according to a predetermined window function. ｃ -1) A motor speed control device according to claim 1 or 2, comprising the element.

4. Regarding a power converter having multiple semiconductor switching elements that converts DC power to AC power and supplies it to an AC motor, The current supplied to the AC motor is detected, The speed of the aforementioned AC motor is detected, When controlling the on / off state of the plurality of semiconductor switching elements, A command current is generated based on the deviation between the input command speed and the detected speed. Based on the deviation between the command current and the detected current, a command voltage is generated. A motor speed control method that calculates, for each control cycle, a command current to make the AC motor's speed track a target speed in a control cycle multiple cycles ahead of the current control cycle, by summing the command current calculated in the previous cycle with a correction current calculated in the current cycle.

5. The motor speed control method according to claim 4, which calculates and outputs a command voltage that makes the deviation between the command current and the detected current zero in the next control cycle from the current control cycle.

6. The command current is calculated as the vector sum of the previous command current and the correction current. The aforementioned plurality of values ​​n ｃ If so, the vector matrix of the correction current is the first to the (n) elements, each of which is weighted according to a predetermined window function. ｃ -1) A motor speed control method according to claim 4 or 5, comprising the element.

Images