JP7465735B2 - Control device - Google Patents

Description

The present invention relates to a control device that controls the torque of an electric motor.

From the viewpoint of improving the maintainability of electric motors and making electric motors smaller and more powerful, there are known motor control devices that are not provided with a speed sensor (i.e., do not use a speed sensor) (so-called speed sensorless electric motor control devices) (see, for example, Patent Document 1). In such control devices, it is generally necessary to estimate the speed (angular velocity) of the electric motor when controlling the torque of the electric motor.

Figure 3 shows an example of the configuration of a conventional control device 100A, which is a so-called speed sensorless motor control device that does not use a speed sensor.

The control device 100A shown in FIG. 3 includes a current detection unit 2, a pick-up control unit 3, an initial value estimation unit 4, a torque control unit 5, a switching unit 6, and a power conversion unit 7.

The current detection unit 2 detects the current vector i flowing through the electric motor 1 and outputs it to the pick-up control unit 3, the initial value estimation unit 4, and the torque control unit 5.

The pick-up control unit 3 receives the current vector i detected by the current detection unit 2, the DC current command I, and the current phase angle command θ. The pick-up control unit 3 generates a pick-up voltage command V0 to make the current i flowing through the motor 1 a DC current according to the DC current command I and the current phase angle command θ, and outputs the voltage command V0 to the initial value estimation unit 4 and the switching unit 6.

The initial value estimation unit 4 receives the current vector i detected by the current detection unit 2 and the pick-up voltage command V0 generated by the pick-up control unit 3. Based on the current vector i and the pick-up voltage command V0, the initial value estimation unit 4 calculates the initial speed ωm0 and initial secondary magnetic flux φ20 of the motor 1 and outputs them to the torque control unit 5. The configuration of the initial value estimation unit 4 will be described in detail later.

The torque control unit 5 receives the current vector i detected by the current detection unit 2, the initial speed ωm0 and initial secondary magnetic flux φ20 of the motor 1 calculated by the initial value estimation unit 4, and a torque control start command SW. When the torque control start command SW is turned on, the torque control unit 5 generates a torque control command V1 for controlling the torque of the motor 1 based on the current vector i, using the initial speed ωm0 and the initial secondary magnetic flux φ20 as initial values, and outputs the generated command to the switching unit 6. In FIG. 3, the torque control start command SW is in the off state when the power supply to the motor 1 is turned on.

The switching unit 6 receives the pick-up voltage command V0 generated by the pick-up control unit 3, the torque control command V1 generated by the torque control unit 5, and the torque control start command SW. In accordance with the control start command SW, the switching unit 6 switches between the pick-up voltage command V0 and the torque control command V1, and outputs the voltage command V ^* to the power conversion unit 7. That is, the switching unit 6 outputs the pick-up voltage command V0 as the voltage command V ^* until the torque control start command SW is turned on, and when the torque control start command SW is turned on, outputs the torque control command V1 as the voltage command V ^* .

The power conversion unit 7 amplifies the voltage command V ^* output from the switching unit 6 and supplies electric power to the electric motor 1 .

In the control device 100A shown in FIG. 3, the pick-up control unit 2 and the initial value estimation unit 4 estimate the initial speed ωm0 and initial secondary magnetic flux φ20 of the motor 1 until the torque control start command SW is turned on. Then, when the torque control start command SW is turned on, the torque control of the motor 1 is performed using the initial speed ωm0 and initial secondary magnetic flux φ20 at the time when the torque control start command SW is turned on as initial values. The timing to turn on the torque control start command SW is determined based on the implementation time of the pick-up control, the state of the initial speed ωm0 and the initial secondary magnetic flux φ20, etc.

Below, we will explain in detail the operation of the initial value estimation unit 4 during pick-up control until the torque control start command SW is turned on.

As shown in FIG. 3, the initial value estimation unit 4 includes a calculation timer 8, a magnetic flux estimation unit 9, a magnetic flux memory 10, a three-point extraction unit 11, a speed estimation unit 12, and an initial magnetic flux estimation unit 13.

The pick-up control start command ST is input to the calculation timer 8. The calculation timer 8 is a timer counter that is cleared to 0 at the edge of the pick-up control start command ST, and outputs the pick-up time t0 to the magnetic flux memory 10, the three-time point extraction unit 11, the speed estimation unit 12, and the initial magnetic flux estimation unit 13.

The magnetic flux estimation unit 9 receives the current vector i detected by the current detection unit 2 and the pick-up voltage command V0 generated by the pick-up control unit 3. The magnetic flux estimation unit 9 calculates the magnetic flux vector φ2r of the motor 1 based on the current vector i and the pick-up voltage command V0 using the following formula (1).

In equation (1), R1, L1, L2, and M are the primary resistance, primary self-inductance, secondary self-inductance, and mutual inductance of motor 1, respectively.

The magnetic flux estimation unit 9 outputs the calculated magnetic flux vector φ2r to the magnetic flux memory 10.

The magnetic flux memory 10 receives the pick-up time t0 generated by the calculation timer 8 and the magnetic flux vector φ2r calculated by the magnetic flux estimation unit 9. The magnetic flux memory 10 stores the magnetic flux vector φ2r, which changes from moment to moment, in the section from time 0 to the pick-up time t0, and outputs it to the three-time point extraction unit 11 as the magnetic flux vector φ(t0).

The three-time point extraction unit 11 receives the picked-up magnetic flux t0 generated by the calculation timer 8 and the magnetic flux vector φ(t0) output from the magnetic flux memory 10. The three-time point extraction unit 11 extracts magnetic flux vectors φ(t00), φ(t01), and φ(t02), which are the magnetic flux vector φ2r at any three times t00, t01, and t02 in the section from time 0 to time t0, and outputs them to the speed estimation unit 12.

The speed estimation unit 12 receives the pick-up time t0 generated by the calculation timer 8 and the three magnetic flux vectors φ(t00), φ(t01), and φ(t02) extracted by the three-time point extraction unit 11. The speed estimation unit 12 calculates the initial speed ωm0 based on the pick-up time t0 and the magnetic flux vectors φ(t00), φ(t01), and φ(t02) using the following equations (2) to (7).

First, the speed estimation unit 12 finds the center R of a circle that passes through the end points of each of the three magnetic flux vectors φ(t00), φ(t01), and φ(t02) using the following equations (2) to (5).

In equations (2) to (5), vector F1 is a vector that indicates the end point of the magnetic flux vector φ(t01) as viewed from the end point of the magnetic flux vector φ(t00). Vector F2 is a vector that indicates the end point of the magnetic flux vector φ(t02) as viewed from the end point of the magnetic flux vector φ(t00). ( _F1A , _F1B ), ( _F2A , _F2B ), and (R _A , R _B ) are the components of vector F1, vector F2, and center R of the circle, respectively.

Next, the speed estimation unit 12 calculates the angle θC between the magnetic flux vector φ(t00) and the magnetic flux vector φ(t02) as viewed from the center R of the circle using equation (6).

In equation (6), the symbol "×" indicates the cross product of vectors, and the symbol "·" indicates the inner product of vectors.

The rotational speed of the electric motor 1 from time t00 to time t02 (e.g., several tens of milliseconds) can be considered to be almost constant. Therefore, the speed estimation unit 12 can calculate the initial speed ωm0 using the following equation (7).

The speed calculation unit 12 outputs the calculated initial speed ωm0 to the torque control unit 5 and the initial magnetic flux estimation unit 13.

The initial magnetic flux estimation unit 13 receives the current vector i detected by the current detection unit 2, the pick-up time t0 generated by the calculation timer 8, and the initial speed ωm0 calculated by the speed calculation unit 12. The initial magnetic flux estimation unit 13 calculates the initial secondary magnetic flux φ20 based on the current vector i and the initial speed ωm0 using the following equation (8).

In equation (8), "j" is the imaginary unit. T2 is the secondary time constant of motor 1, and is calculated by L2/R2 using secondary resistance R2. φxx is the residual magnetic flux at the start of pick-up control.

JP 2000-287491 A

There is room for improvement in the pick-up control described above, as explained below.

When an error occurs in the primary resistance R1 due to temperature fluctuations of the motor 1, the error is integrated into the magnetic flux vector φ2r, which is the output value of the magnetic flux estimation unit 9, by the integral term of equation (1), as shown in equation (11) described below. As a result, errors occur in the magnetic flux vectors φ(t00), φ(t01), and φ(t02), which are the output of the three-time point extraction unit 11. In addition, errors also occur in the initial speed ωm0 and initial magnetic flux φ20 calculated by equations (2) to (8) based on the magnetic flux vectors φ(t00), φ(t01), and φ(t02).

In the following, it is assumed that the primary resistance R1 deviates from the true value R1real by an error ΔR as shown in equation (9).
R1real=R1+ΔR ...Equation (9)

In this case, the real magnetic flux vector φ2real is calculated using the following equation (10):

When equation (9) is substituted for R1real in equation (10), since the current vector i is constant, equation (10) can be expressed as the following equation (11) using equation (1).
φ2r=φ2real+ΔR·i·t Equation (11)

In equation (11), t is the calculation time of the magnetic flux vector φ2r. Also, L2 ≒ M.

As shown in equation (11), if an error occurs in the primary resistance R1, the error is integrated into the magnetic flux vector φ2r, which is the output value of the magnetic flux estimation unit 9, by the integral term in equation (1).

In this way, when an error exists in the primary resistance R1, an error occurs in the initial speed ωm0 and the initial secondary magnetic flux φ20 estimated by the initial value estimation unit 4. If an error occurs in the initial speed ωm0 and the initial secondary magnetic flux φ20, which are the initial values for the torque control by the torque control unit 5, the accuracy of the torque control of the electric motor 1 may decrease.

The objective of the present invention, which has been made in consideration of the above problems, is to provide a control device that can improve the accuracy of torque control of an electric motor.

In order to solve the above problem, the control device according to the present invention is a control device for an electric motor, and includes a current detection unit that detects a current vector flowing through the electric motor, a pick-up control unit that outputs a pick-up voltage command for passing a direct current through the electric motor based on the current vector, a magnetic flux estimation unit that calculates a magnetic flux vector of the electric motor based on the current vector and the pick-up voltage command, an intermediate vector extraction unit that uses the magnetic flux vector to extract intermediate vectors for multiple time points after the starting point time based on the magnetic flux vector at a predetermined starting point time, the magnetic flux vector at a predetermined time point after the starting point time, and the magnetic flux vector at an intermediate time point between the starting point time and the predetermined time point, a speed estimation unit that calculates an initial speed of the electric motor based on the intermediate vectors extracted for the multiple time points, and an initial magnetic flux estimation unit that calculates an initial secondary magnetic flux of the electric motor based on the initial speed and the current vector.

Furthermore, in the control device according to the present invention, the intermediate vector extraction unit calculates an intermediate vector φc(tn) at the predetermined time point by the following equation (12), where the magnetic flux vector at the start time point is φ(ts), the magnetic flux vector at the predetermined time point is φ(tn), and the magnetic flux vector at an intermediate time point between the start time point and the predetermined time point is φ((ts+tn)/2):
φc(tn)=2*[φ((ts+tn)/2)-φ(ts)]
−[φ(tn)−φ(ts)] ... Equation (12)
The speed estimation unit calculates the initial speed ωm0 by the following equation (13), where the vector angle of the intermediate vector φc(t1) at a first time point t1 after the start point time is θ(t1) and the vector angle of the intermediate vector φc(t2) at a second time point t2 after the start point time that is different from the first time point t1 is θ(t2).
ωm0=2*(θc1-θc2)/(t1-t2) ...Equation (13)

The control device according to the present invention can improve the accuracy of torque control of the electric motor.

FIG. 2 is a diagram illustrating a configuration example of a control device according to an embodiment of the present invention. 2 is a diagram showing an intermediate vector calculated by an intermediate vector extraction unit shown in FIG. 1; FIG. 1 is a diagram illustrating an example of the configuration of a conventional control device.

The following describes the embodiment of the present invention with reference to the drawings.

Figure 1 is a diagram showing an example of the configuration of a control device 100 according to one embodiment of the present invention. The control device 100 according to this embodiment controls the torque of an electric motor 1. In Figure 1, the same components as those in Figure 3 are given the same reference numerals, and the description will be omitted.

The control device 100 shown in FIG. 1 includes a current detection unit 2, a pick-up control unit 3, a torque control unit 5, a switching unit 6, a power conversion unit 7, a calculation timer 8, a magnetic flux estimation unit 9, a magnetic flux memory 10, an initial magnetic flux estimation unit 13, an intermediate vector extraction unit 101, and a speed estimation unit 102. That is, the control device 100 according to this embodiment differs from the control device 100A shown in FIG. 3 in that the three-time point extraction unit 11 and the speed estimation unit 12 are changed to the intermediate vector extraction unit 101 and the speed estimation unit 102, respectively.

The intermediate vector extraction unit 101 calculates intermediate vectors at multiple points in time based on the magnetic flux vector φ2r and outputs them to the speed estimation unit 102. The intermediate vectors are described in more detail below.

The intermediate vector extraction unit 101 calculates an intermediate vector φc(tn) at a predetermined time tn based on a magnetic flux vector φ(ts) at a predetermined start time ts, a magnetic flux vector φ(tn) at a predetermined time tn after the start time ts, and a magnetic flux vector φ((ts+tn)/2) at an intermediate time ((ts+tn)/2) between the start time ts and the predetermined time tn. More specifically, the intermediate vector extraction unit 101 calculates the intermediate vector φc(tn) by the following formula (14).
φc(tn)=2*[φ((ts+tn)/2)-φ(ts)]
−[φ(tn)−φ(ts)] ... Equation (14)

The intermediate vector extraction unit 101 calculates the intermediate vector φc(tn) described above for multiple time points.

In the following, an example will be described in which the intermediate vector extraction unit 101 calculates an intermediate vector φc(t1) at time t1 (first time) after the start time ts, and an intermediate vector φc(t2) at time t2 (second time) after the start time ts that is different from time t1. In the following, the intermediate time between the start time ts and time t1 is defined as tA, and the magnetic flux vector at time tA is defined as φ(tA). In addition, the intermediate time between the start time ts and time t2 is defined as tB, and the magnetic flux vector at time tB is defined as φ(tB).

The intermediate vector extraction unit 101 calculates the intermediate vector φc(t1) by the following equation (15), and calculates the intermediate vector φc(t2) by the following equation (16).
φc(t1)=2*[φ(tA)-φ(ts)]-[φ(t1)-φ(ts)]
...Equation (15)
φc(t2)=2*[φ(tB)-φ(ts)]-[φ(t2)-φ(ts)]
...Equation (16)

The intermediate vector φc(t1) and the intermediate vector φc(t2) are shown in FIG. 2. As shown in FIG. 2, time t1 is an arbitrary time on the trajectory (circle) of the magnetic flux vector φ2r after the starting time ts. The intermediate vector φc(t1) at time t1 is obtained by adding the magnetic flux vector φ(ts) to 2*(magnetic flux vector φ(tA)-magnetic flux vector φ(ts)) and then adding the magnetic flux vector φ(t1). The intermediate vector φc(t2) at time t2 is obtained by adding the magnetic flux vector φ(ts) to 2*(magnetic flux vector φ(tB)-magnetic flux vector φ(ts)) and then adding the magnetic flux vector φ(t2).

The above-mentioned intermediate vector φc(tn) (intermediate vectors φc(t1), φc(t2)) is not affected by the error ΔR in the primary resistance R1 of the motor 1. The reason for this will be explained below.

It is assumed that the primary resistance R1 deviates from the true value R1real by an error ΔR. In this case, from equation (11), the magnetic flux vectors at the start time ts, time point t1, and time point tA are expressed by the following equations (17) to (19).
φ(ts)=φ2real(ts)+ΔR·i·ts Equation (17)
φ(t1)=φ2real(t1)+ΔR·i·t1 ... Equation (18)
φ(tA)=φ2real(tA)+ΔR·i·tA Equation (19)

Considering equation (15), the intermediate vector φc(t1) is expressed by the following equation (20).
φc(t1)=2*[φreal(tA)−φreal(ts)]
−[φreal(t1)−φreal(ts)]
+ΔR i [2 * (tA-ts)-(t1-ts)] ... Equation (20)

Since tA = (t1 + ts)/2, the ΔR term in equation (20) is 0. In other words, it can be seen that the intermediate vector φc(t1) is not affected by the error ΔR. Similarly, it can be seen that the intermediate vector φc(t2) is not affected by the error ΔR.

The speed estimation unit 102 calculates the initial speed ωm0 of the motor 1 based on the intermediate vector φc(tn) extracted for multiple time points. The calculation of the initial speed ωm0 by the speed estimation unit 102 will be explained in more detail using the above-mentioned time points t1 and t2 as examples.

If the components of the intermediate vector φc(t1) at time t1 are (FCT1A, FCT1B), the speed estimation unit 102 calculates the vector angle θc1 of the intermediate vector φc(t1) by the following equation (21).
θc1=tan ⁻¹ (FCT1B/FCT1A) (FCT1A≧0)
θc1=π+tan ⁻¹ (FCT1B/FCT1A) (FCT1A<0)
...Equation (21)

Furthermore, if the components of the intermediate vector φc(t2) at time t2 are (FCT2A, FCT2B), the speed estimation unit 102 calculates the vector angle θc2 of the intermediate vector φc(t2) by the following equation (22).
θc2=tan ⁻¹ (FCT2B/FCT2A) (FCT2A≧0)
θc2=π+tan ⁻¹ (FCT2B/FCT2A) (FCT2A<0)
...Equation (22)

Vector angle θc1-vector angle θc2 is the angle of the magnetic flux vector rotated between time tA and time tB. Since the time between time tA and time tB is half the time from time t1 to time t2, the speed estimation unit 102 calculates the initial speed ωm0 by the following equation (23).
ωm0=2*(θc1-θc2)/(t1-t2) ... Equation (23)

The initial speed ωm0 calculated by equation (23) is calculated from the intermediate vectors φc(t1) and φc(t2), and as described above, the intermediate vectors φc(t1) and φc(t2) are not affected by the error ΔR in the primary resistance R1 of the motor 1. Therefore, in this embodiment, the initial speed ωm0 of the motor 1 can be calculated without being affected by the error ΔR in the primary resistance R1.

As described above, the initial magnetic flux estimation unit 13 calculates the initial secondary magnetic flux φ20 using equation (8) based on the current vector i and the initial speed ωm0. Since the initial speed ωm0 is not affected by the error ΔR of the primary resistance R1, the initial secondary magnetic flux φ20 can also be calculated without being affected by the error ΔR of the primary resistance R1.

In this way, in this embodiment, the initial speed ωm0 and the initial secondary magnetic flux φ20 of the motor 1 can be calculated without being affected by the error ΔR of the primary resistance R1. Therefore, it is possible to improve the accuracy of the torque control of the motor 1 using the initial speed ωm0 and the initial secondary magnetic flux φ20.

There are limitations to the calculation of the initial speed ωm0 using equations (21) to (23). If the trajectory of the magnetic flux vector φ2r rotates more than once, it becomes impossible to distinguish between low speed and high speed, or forward and reverse rotation. For this reason, it is necessary to select times t1 and t2 within one revolution of the trajectory of the magnetic flux vector φ2r relative to the start time ts.

Regarding the above-mentioned formulas (15) and (16), the right-hand side may be multiplied by 1/2 as shown in the following formulas (15') and (16'). It is clear that the same effect can be obtained in this case. The multiple of the right-hand side may be any positive number other than 1/2, and may not be 0.
φc(t1)=[φ(tA)-φ(ts)]-[φ(t1)-φ(ts)]/2
... Formula (15')
φc(t2)=[φ(tB)-φ(ts)]-[φ(t2)-φ(ts)]/2
... Formula (16')

The speed estimation unit 102 may also calculate the initial speed ωm0 multiple times by performing calculations similar to those of equations (15), (16), and (21) to (23) using a set of two time points that are different from time points t1 and t2 by at least one time point. The speed estimation unit 102 may also output the average value of the initial speed ωm0 calculated multiple times or a value that has passed through a first-order lag filter as the initial speed ωm0.

In this embodiment, the control device 100 includes a current detection unit 2 that detects a current vector i flowing through the motor 1, a pick-up control unit 3 that outputs a pick-up voltage command V1 that causes a DC current to flow through the motor 1 based on the current vector i, a magnetic flux estimation unit 9 that calculates a magnetic flux vector φ2r of the motor 1 based on the current vector i and the pick-up voltage command V1, and a magnetic flux estimation unit 10 that calculates a magnetic flux vector φ(ts) at a predetermined start time ts and a magnetic flux vector φ(ts) at a predetermined time point tn after the start time ts using the magnetic flux vector φ2r. The system includes an intermediate vector extraction unit 101 that extracts intermediate vectors φc(tn) for multiple time points based on magnetic flux vector φ(tn) and magnetic flux vector φ((ts+tn)/2) at an intermediate time point between start time ts and a predetermined time point tn, a speed estimation unit 102 that calculates an initial speed ωm0 of the motor 1 based on the intermediate vectors φc(tn) extracted for multiple time points, and an initial magnetic flux estimation unit 13 that calculates an initial secondary magnetic flux φ20 of the motor 1 based on the initial speed ωm0 and the current vector i.

The intermediate vector φc(tn) based on the flux vector φ(ts), the flux vector φ(tn), and the flux vector φ((ts+tn)/2) is not affected by the error ΔR of the primary resistance R1 of the electric motor 1. Therefore, by calculating the initial speed ωm0 based on the intermediate vector φc(tn), the initial speed ωm0 can be calculated without being affected by the error ΔR of the primary resistance R1. In addition, by calculating the initial secondary magnetic flux φ20 of the electric motor 1 based on the initial speed ωm0 calculated based on the intermediate vector φc(tn), the initial secondary magnetic flux φ20 can be calculated without being affected by the error ΔR of the primary resistance R1. Therefore, since the initial speed ωm0 and the initial secondary magnetic flux φ20 used in the torque control of the electric motor 1 can be calculated without being affected by the error ΔR of the primary resistance R1, the accuracy of the torque control of the electric motor 1 can be improved. Therefore, for example, in an electric vehicle equipped with the electric motor 1, re-acceleration from a coasting state and braking from a coasting state are possible.

The above-described embodiments have been described as representative examples, but it will be apparent to those skilled in the art that many modifications and substitutions are possible within the spirit and scope of the present invention. Therefore, the present invention should not be construed as being limited by the above-described embodiments, and various modifications and changes are possible without departing from the scope of the claims.

REFERENCE SIGNS LIST 1 Electric motor 2 Current detection unit 3 Pick-up control unit 5 Torque control unit 6 Switching unit 7 Power conversion unit 8 Calculation timer 9 Magnetic flux estimation unit 10 Magnetic flux memory 13 Initial magnetic flux estimation unit 100 Control device 101 Intermediate vector extraction unit 102 Speed estimation unit

Claims (2)

A control device for an induction machine ,
A current detection unit that detects a current vector flowing through the induction machine ;
a pick-up control unit that outputs a pick-up voltage command for causing a DC current to flow through the induction machine based on the current vector;
a magnetic flux estimator that calculates a magnetic flux vector of the induction machine based on the current vector and the pick-up voltage command;
an intermediate vector extracting unit that uses the magnetic flux vector to extract intermediate vectors for a plurality of time points after the start time based on the magnetic flux vector at a predetermined start time, the magnetic flux vector at a predetermined time point after the start time, and the magnetic flux vector at an intermediate time point between the start time and the predetermined time point;
a speed estimator that calculates an initial speed of the induction machine based on the intermediate vectors extracted for the plurality of time points;
an initial magnetic flux estimator that calculates an initial secondary magnetic flux of the induction machine based on the initial speed and the current vector. 2. The control device according to claim 1,
The intermediate vector extraction unit
If the magnetic flux vector at the start time is φ(ts), the magnetic flux vector at the predetermined time point is φ(tn), and the magnetic flux vector at an intermediate time point between the start time point and the predetermined time point is φ((ts+tn)/2), an intermediate vector φc(tn) at the predetermined time point is calculated by the following formula (1):
φc(tn)=2*[φ((ts+tn)/2)-φ(ts)]
−[φ(tn)−φ(ts)] ... Equation (1)
The speed estimation unit calculates the initial speed ωm0 by the following equation (2), where θ(t1) is a vector angle of the intermediate vector φc(t1) at a first time point t1 after the start point time, and θ(t2) is a vector angle of the intermediate vector φc(t2) at a second time point t2 after the start point time that is different from the first time point t1.
ωm0=2*(θc1-θc2)/(t1-t2) ... Equation (2)

Images