JPH07118927B2 - Torque control method for reluctance motor - Google Patents

Description

The present invention relates to a torque control method for a reluctance motor, and more specifically, it relates to a torque control method for a reluctance motor, in which torque generated with an increase in rotation speed is irrespective of a torque command value. The present invention relates to a novel torque control method that reduces the decrease.

<Prior Art and Problems to be Solved by the Invention> Conventionally, as a mechanism for driving each axis of an industrial robot, a motor that rotates at high speed and a reduction mechanism having a large reduction ratio are generally adopted. , The shaft drive torque is increased.

However, when such a mechanism is adopted, the effect of backlash cannot be completely eliminated because the reduction mechanism is generally composed of a plurality of gears, and high positioning accuracy is not achieved. It cannot be applied to required industrial robots. Further, if the reduction ratio is set large in order to obtain a large torque, high speed operation cannot be performed.

In order to solve such a problem, it is considered to use the direct drive method that directly transmits the rotational force of the motor to each axis, and if you use the direct drive method, you can obtain a sufficiently large driving force. Reluctance motors are attracting attention as motors that can achieve this. In this reluctance motor, torque is generated based on the difference between the energy supplied to the motor and the energy stored inside the motor. Specifically, it is known that the generated torque τ of a three-phase reluctance motor has a significantly smaller mutual inductance than self-inductance, so τ = τ (θm, ia, ib, ic) ≈ (1 / 2) (dLa / dθm) ia ² + (1/2) (dLb / dθm) ib ² + (1/2) (dLc / dθm) ic ² (where i is the current value, L is the self-inductance, subscript a,
b and c represent phases, and θm represents rotor position). However, since the generated torque pulsates due to the mechanical structure of the reluctance motor itself, it is necessary to greatly reduce the generated torque pulsation when incorporating it into an industrial robot. In order to realize it, the supply current of each phase is controlled (Japanese Patent Laid-Open No. 2000-242242).
63-35194 and "Theoretical analysis of megatorque motor and its torque control method" Tanaka et al., IEEJ Study Group, SPC-87.
-14, PP.1-10, 1987).

However, even if the desired supply current pattern for the reluctance motor can be obtained in this way, the actual winding current does not follow the command value because the reluctance motor itself has a large inductance, There is a problem in that the torque cannot be generated. Specifically, the self-inductances La, Lb, Lc of each phase of the three-phase reluctance motor are (However, θe is an electrical angle, and L0 and L1 are inductances that change under the influence of magnetic saturation, respectively).

Therefore, the self-inductance La changes periodically within the range of L0-L1 and L0 + L1 as shown in FIG. The same applies to the other self-inductances Lb and Lc.

Also, as for the command current waveform when the motor is stopped, looking at the case where the current is passed at a position where torque can be efficiently generated, that is, a position where the inductance change is large,
Considering the change curves of the self-inductances La, Lb, Lc, ia = Ip in the range of π / 6 ≦ θe ≦ 5π / 6, and ia = 0 in other ranges.
(See FIG. 17), ib = Ip in the range of 5π / 6 ≦ θe ≦ 3π / 2, ib = in other ranges
0, ic = Ip in the range of θe ≤ π / 6 or 3π / 2 ≤ θe, and ic = 0 in other ranges. However, due to the large self-inductance, However, there is a problem in that a sufficient generated torque cannot be achieved especially in a high speed rotation state.

Figure 18 A, B, C is a diagram showing the actual winding current and the torque generated for one phase with respect to the current command value, showing the case where the reluctance motor rotation speed is 0.1rps, 0.5rps, 1.0rps, respectively. ing. As is clear from this figure, as the rotational speed increases, the delay in the rise and fall of the winding current increases and the generated torque also decreases. In particular, in the case of 1.0 rps, the winding current value does not reach the command current value, and therefore the generated torque is significantly small.

Although only the changes in the winding current and the generated torque with the increase of the rotation speed have been described above, even if the generated torque command value increases, the winding current rises even if the rotation speed is the same, As the trailing delay increases, the generated torque also decreases.

Therefore, even if an appropriate supply current pattern is obtained, the winding current will deviate considerably from the supply current pattern, and the reluctance motor cannot be rotated as intended.

<Objects of the Invention> The present invention has been made in view of the above problems,
It is an object of the present invention to provide a torque control method for a reluctance motor that can effectively compensate for a deviation of a winding current that changes depending on a rotation speed and a torque command value and increase a generated torque.

<Means for Solving the Problems> A torque control method for a reluctance motor according to the present invention to achieve the above object is to supply currents having different electrical angles of 2π / n to each phase of an n-phase reluctance motor. In the torque control method described above, based on the reluctance motor rotation speed and the torque command value, the deviation angle of the excitation timing is obtained corresponding to the energization start time and the energization end time, and each phase is shifted at the timing shifted by the deviation angle. This is a method of supplying an electric current.

In these cases, it is preferable that the deviation angle of the excitation timing be obtained by reading the deviation angle stored in advance in a two-dimensional table of the rotation speed and the torque command value and performing the interpolation calculation.

In addition, the reluctance motor is a three-phase reluctance motor, the excitation section for any phase is divided into three sections, and the generated torque obtained from the measured data is obtained for the single section in which only the relevant phase is excited. Calculate the supply current pattern that makes the generated torque constant based on the characteristics,
An arbitrary supply current pattern is set for one of the composite sections in which the relevant phase and the other phase are simultaneously excited, and the torque generated when the other phase is excited in the arbitrary supply current pattern for the other of the composite section. The constant supply current pattern is calculated based on the generated torque characteristics obtained from the measurement data, and the supply current of each phase is calculated based on the continuous supply current pattern, and the calculated value of the current is calculated. Is preferably supplied to each phase at a timing shifted by a shift angle.

Further, the reluctance motor is a three-phase reluctance motor, the excitation section for any phase is divided into three sections, and the independent section in which only the corresponding phase is excited is based on the preset loss. Calculate the supply current pattern that maximizes the generated torque, and generate one based on the generated torque characteristics obtained from the measured data and the preset loss for one of the complex sections in which the relevant phase and the other phase are simultaneously excited. The supply current patterns of the corresponding phase and other phases that maximize the torque are calculated, and the supply current patterns of the other phases are applied to the other of the complex sections [I] and [III] to continuously connect the supply current patterns. It is preferable to calculate the supply current of each phase based on the pattern that was set and supply the calculated value of the current to each phase at the timing shifted by the deviation angle. .

Furthermore, the reluctance motor is a three-phase reluctance motor, the excitation section for any phase is divided into three sections, and the single section in which only the corresponding phase is excited is obtained from the measured data. Calculate the supply current pattern that makes the generated torque constant based on the torque characteristics, set an arbitrary supply current pattern for one of the composite sections in which the relevant phase and the other phase are simultaneously excited, and set the other for the other section. Is calculated based on the generated torque characteristics obtained from the measured data, the supply current pattern that makes the generated torque constant when the other phase is excited with an arbitrary supplied current pattern, and the supply current pattern is made continuous. One pattern is obtained, and the single section [II] in which only the relevant phase is excited is generated based on the preset loss. The supply current pattern that maximizes the torque is calculated, and for one of the complex sections [I] and [III] in which the corresponding phase and the other phase are simultaneously excited, the generated torque characteristics obtained from the measured data and preset values are set. Based on the loss, the supply current patterns of the relevant phase and other phases that maximize the generated torque are calculated, and the composite section [I]
Regarding the other of [III], the supply current pattern of the other phase is applied to obtain a pattern in which the supply current pattern is continuous, and the first pattern or the second pattern is selected to obtain the selected pattern. It is preferable to calculate the supply current value of each phase based on the above, and supply the calculated value of the current to each phase at a timing shifted by the shift angle.

<Operation> According to the torque control method of the above invention, when the generated torque is controlled by supplying currents having different electrical angles of 2π / n to each phase of the n-phase reluctance motor, the rotation of the reluctance motor is controlled. Based on the speed and torque command values, a predetermined calculation is performed, or the measured values that have already been obtained are read to obtain the deviation angles of the excitation timings corresponding to the energization start point and the energization end point, respectively. Since the current is supplied to each phase at the timing shifted by the deviation angle, the deviation between the supply current and the actual winding current can be greatly suppressed, and it occurs with the increase of the rotation speed and the torque command value. The decrease in torque can be greatly suppressed, and the generated torque can be increased.

In these cases, the deviation angle of the excitation timing is
If it can be obtained by reading the displacement angle stored in advance in the two-dimensional table of the rotation speed and the torque command value and performing the interpolation calculation, it is possible to reduce the number of actually measured values and only perform the simple interpolation calculation. Thus, the deviation between the supply current and the actual winding current can be greatly suppressed, and the reduction in the generated torque due to the increase in the rotation speed and the torque command value can be greatly suppressed.

Further, according to the torque control method of the third aspect of the invention, it is possible to obtain the supply current waveform that greatly reduces the torque pulsation based on a small number of measured values, and shift the obtained supply current waveform by the shift angle. Since it is supplied to each phase at the timing,
The deviation between the supply current and the actual winding current can be greatly suppressed, the decrease in the generated torque due to the increase in the rotation speed and the torque command value can be greatly suppressed, and eventually the torque pulsation. It is possible to greatly suppress the increase of

Further, according to the torque control method of the fourth aspect of the present invention, it is possible to obtain a supply current waveform that efficiently generates torque based on a small actual measurement value, and to obtain each of the obtained supply current waveforms at a timing shifted by a deviation angle. Since it is supplied to each phase, the deviation between the supply current and the actual winding current can be greatly suppressed, and the decrease in the generated torque due to the increase of the rotation speed and the torque command value can be greatly suppressed. Therefore, the reduction of the generated torque can be greatly suppressed over the entire range.

Furthermore, according to the torque control method of the fifth aspect of the present invention, a supply current waveform that greatly reduces the torque pulsation and a supply current waveform that efficiently generates torque are obtained based on a small number of measured values. Since one of the supply current waveforms is selected and supplied to each phase at a timing shifted by the deviation angle, the deviation between the supply current and the actual winding current can be greatly suppressed. Therefore, the reduction of the generated torque due to the increase of the rotation speed and the torque command value is greatly suppressed,
It is possible to select a state in which an increase in torque pulsation can be greatly suppressed and a state in which a decrease in generated torque can be greatly suppressed over the entire range.

In this case, the selection of both supply current patterns may be performed based on the magnitude of the torque command value. Generally, when high speed operation is required, the torque command value is large,
On the contrary, when the positioning operation is urgently performed, the torque command value is small, so that the optimum pattern for the torque control of the reluctance motor is selected. Further, both supply current patterns may be selected based on the magnitude of the position deviation or the magnitude of the speed deviation, and either pattern may be selected depending on how much the actual operation follows the target. Is selected.

The supply current pattern will be described in detail.

Torque generated by a three-phase reluctance motor τ (θm, ia, ib,
ic) (where i is the current value, subscripts a, b and c are phases, and θm is the rotor position), τ (θm, ia, ib, ic) = τa (θm, ia, ib , ic) + τb (θm, ia, ib, ic) + τc (θm, ia, ib, ic), but since the reluctance motor has a remarkably small mutual inductance, τa (θm, ia, ib, ic) ) ≈ τa (θm, ia, 0,0) τb (θm, ia, ib, ic) ≈τb (θm, 0, ib, 0) τc (θm, ia, ib, ic) ≈τc (θm, 0, 0, ic), and as a result, the generated torque is also τ (θm, ia, ib, ic) = τa (θm, ia, 0,0) + τb (θm,
It can be approximated as 0, ib, 0) + τc (θm, 0,0, ic). Therefore, the required actual measurement value is only required for the one-phase excitation, and the three-dimensional data can be converted into one-dimensional data.

Then, if the number of pole pairs is p and the electrical angle is θe, the characteristics between the poles are the same, so τ (θe, ia, ib, ic) = τ (θe + 2nπ / p, ia, ib, ic) ( However, n is an arbitrary integer, and the actual measurement value required is only within the range of the electrical angle of 0 to 2π, and can be reduced to 1 / p of the number of data in the entire rotation range.

Further, since the characteristics between the phases are equal, τa (θe, ia, ib, ic) = τb (θe-2π / 3, ia, ib, ic) = τc (θe + 2π / 3, ia, ib, ic) Therefore, the actual measurement value required is only for one phase, and can be reduced to 1/3.

Further, as shown in FIG. 12, when the characteristics of each phase are symmetric, τ (θe, ia, ib, ic) = −τ (−θe, ia, ib, ic) The measured value is only for the symmetrical one side section,
It can be reduced to 1/2.

Therefore, by taking these into consideration, the number of required actual measurement values can be significantly reduced, and the measured values are discretely obtained, and interpolation calculation is performed based on the measured values to reduce torque pulsation. If the supply current pattern for causing the generated torque and the supply current pattern for maximizing the generated torque are obtained without increasing the loss, the number of actually measured values required can be further reduced.

Then, the required output torque τ * τ * = τ (θm, ia, ib, ic) (τ * is a constant value) is the supply current ia, ib, ic of each phase required to obtain the former supply current. By obtaining it based on the pattern, the torque control of the reluctance motor can be performed in a state where the torque fluctuation is extremely small, that is, the torque pulsation is extremely small.

If the loss is set in advance, ia ² + ib ² + ic ² = Ip ² (where Ip is a constant determined based on the preset loss) is satisfied, and each phase is obtained by independent excitation. By obtaining the supply currents ia, ib, ic for each phase required to maximize the sum of the generated torques τa, τb, τc based on the latter supply current pattern, the reluctance motor can be operated with high torque generation efficiency. Can be rotated at high speed.

Although only the three-phase reluctance motor has been described above in detail, the same applies to a reluctance motor having an arbitrary number of phases.

<Example> Hereinafter, detailed description will be given with reference to the accompanying drawings illustrating an example.

FIG. 6 is a block diagram showing an example of the electrical configuration of a reluctance motor control device that implements the torque control method of the present invention. It receives a torque command value supplied from a host controller (not shown) as an input and a predetermined torque. To calculate the supply current value for each phase by reading the supply current value from the pattern table (2) that stores the supply current pattern for one phase required to generate ) And an excitation timing change unit (5) that calculates a deviation angle of the excitation timing that requires the calculated supply current value and outputs the supply current value in a state of being displaced by the deviation angle, and the excitation timing is changed. Also, a power conversion unit (3) that converts the supply current value for each phase into the supply power for each phase, and the power converted by the power conversion unit (3). It has a three-phase reluctance motor (4) that rotates to generate a commanded torque when supplied, and outputs from a rotor position sensor (4a) incorporated in the three-phase reluctance motor (4). The detected rotor position detection signal is directly supplied to the arithmetic unit (1), and the arithmetic unit (1) generates a read address and supplies it to the pattern table (2). The rotor position detection signal is converted into a speed signal by the position-speed conversion unit (6) and supplied to the excitation timing changing unit (5). The torque command value is also supplied to the excitation timing changing section (5).

In the pattern table (2), a supply current pattern indicating a change in the supply current for obtaining a predetermined generated torque is stored as a calculation result based on an actual measurement value, and when the generated torque is changed. A plurality of supply current patterns corresponding to are stored (see FIG. 4).

The excitation timing changing unit (5) stores the phase advance angle with respect to the rotation speed when a predetermined torque command value is given, based on the actually measured value, and when the torque command value is changed. A plurality of corresponding phase lead angle patterns are stored (see FIG. 5).

FIG. 1 is a flow chart showing an embodiment of the torque control method of the present invention. In step, only the a phase of the three-phase reluctance motor is DC-excited to operate the reluctance motor to measure the generated torque pattern. , Calculating the phase a supply current pattern based on the generated torque pattern and storing it in the pattern table (2), and measuring the phase advance angle with respect to the rotation speed in correspondence with each of the plurality of torque command values in step,
In step, the plurality of obtained phase lead angle patterns are stored in the excitation timing changing section (5) to complete the initial setting. After that, wait until the torque command value τ * is supplied in the step, and then in the step, the torque command value τ * and the rotor position sensor (4a)
The position θ detected by is read, and in step, the speed dθ / dt is calculated based on the read position θ. Then, in step, the torque command value τ
* And phase lead angle θa determined based on velocity dθ / dt
And newly set θ = θ + θa. Then, in step, the torque command value τ * and the supply current ia corresponding to the new θ are calculated, and in step the position θ is set to 2
Shift by π / 3, calculate the supply current ib in the same way as in step, and set position θ in step.
Is further shifted by 2π / 3, and in step, the supply current ic is calculated in the same manner as in step. Then, in the step, by supplying the calculated supply currents ia, ib, ic to the reluctance motor (4), respectively, the reduction of the generated torque can be greatly controlled.

That is, the supply currents ia, ib and ic are respectively π / 6 ≦ θe ≦
Ia = Ip in the range of 5π / 6, ia = 0 in other ranges, ib = Ip in the range of 5π / 6 ≦ θe ≦ 3π / 2, ib = in other ranges
If the supply current pattern is stored in the pattern table (2) such that ic = Ip in the range of 0, θe ≦ π / 6 or 3π / 2 ≦ θe, and ic = 0 in other ranges If the angular supply currents ia, ib, ic are supplied without shifting the phase, the torque command value τ *
Even if is constant, the actual winding current changes according to the rotation speed as shown in FIGS. 18A, 18B and 18C, and the generated torque also decreases as the winding current changes.

However, if the above torque control method is applied, only the phase of the supply current is advanced by θa (see Fig. 2), so that the actual winding current rises sufficiently even when the rotation speed is 1.0rps. (See FIG. 3A), the generated torque can be greatly increased compared to FIG. 18C (see FIG. 3B).
Although FIG. 3 shows only the case where the generated torque is most significantly reduced, the decrease of the generated torque can be suppressed in the same manner at other rotation speeds, and further, the same rotational speed is obtained. However, by reducing the deviation angle in accordance with the magnitude of the torque command value τ *, it is possible to similarly suppress the decrease in the generated torque.

Although not shown in the above flow chart, as shown in FIG. 14, the phase advance angle θas on the energization start side of the supply current pattern is changed to the phase advance angle θae on the energization end side.
It is set larger than. With this setting, the generated torque can be increased as shown by the solid line in FIG. The alternate long and two short dashes line in FIG. 15 is θas = θae
The generated torque in the case of is shown.

7A, 7B and 7C show other examples of the supply current patterns stored in the pattern table (2), respectively.
Is a supply current pattern suitable for remarkably reducing torque pulsation, FIG. 9B is a supply current pattern suitable for generating torque most efficiently, and FIG. 7C is a torque command value τ.
When * is small, the torque pulsation is remarkably reduced, and when the torque command value τ * is large, the supply current pattern is suitable for efficiently generating torque.

Therefore, even when the pattern table (2) in which the above-mentioned supply current pattern is stored is adopted, the phase shift between the supply current and the actual winding current due to the self-inductance is greatly suppressed, and each of the above-mentioned each Torque control of the reluctance motor (4) suitable for the purpose can be performed. However, in this case, the torque command value τ
When calculating the supply current ia corresponding to *, for example, τ1
The four points where <τ * <τ2 and θ1 <θ <θ2 are read from the pattern table (2), and the supply current is calculated by performing an interpolation operation based on the read four supply current values, At the obtained supply current ia
Should be supplied to the reluctance motor (4).

FIG. 8 is a flowchart showing a method of calculating the supply current pattern of the a phase (see FIG. 7A) based on the remarkably small measured value, and in the step, the electrical angle is π / 3 to
For the section of 2π / 3 (refer to section II in FIG. 9B) τ * = τa (θc, ia, 0,0) (where τ * is a preset predetermined value), each electrical angle θe Find the supply current ia for. Specifically, based on a plurality of measured values, FIG.
As shown in, the supply current ia is obtained. However, since the generated torque pattern is only obtained discretely, for example, when the electrical angle is θe and the supply currents are ia1 and ia2, respectively. The generated torques are τa1 and τa2,
Moreover, if τa1 <τ * <τa2, then the supply current ia (θ
e) = ia1 + {(τ * −τa1) / (τa2-τa1)} (i
a2-ia1). However, instead of the linear interpolation calculation as described above, a higher-order interpolation calculation may be performed to improve the approximation accuracy.

Next, in the step, τ * = τa (θe, ia, 0,0) + τa (θe + 2π / 3,0,0, ic) for the section where the electrical angle is 0 to π / 3 (see section I in FIG. 9B). ), The supply current ia, ic for each electrical angle θe, θe + 2π / 3 is calculated. However, since there are many supply currents ia, ic that satisfy the above relationship, they cannot be uniquely defined. Therefore, specifically, for example, the pattern of the supply current ia in the above section I is a linear pattern {ia = 0 when θ = θ0 as shown in FIG. 10B and ia = I1 when θ = θ1. (However, I1 is a linear pattern that satisfies τ * = τa (θ1, I1,0,0))}, and the supply current ic in section I and the supply current ic based on the above relation
(See FIG. 10C) may be calculated. And, in this case as well, it goes without saying that the interpolation calculation is performed as necessary.

Then, in step, supply current in section I
By adopting the pattern of ic as it is as the pattern of the supply current ia in the section III, as shown in FIG. 10D, the pattern of the supply current ia over the entire range of the electrical angle of 0 to π can be obtained. However, FIG. 9B corresponds to the case where the torque command value τ * is negative, and the torque command value τ *
When is greater than or equal to 0, the phase supply currents ia, ib, ic may be calculated according to the classification shown in FIG. 9A.

FIG. 11 is a flowchart showing a method of calculating the supply current pattern of the a phase (see FIG. 7B) based on the remarkably small measured value.
The supply current ia is determined so that ia = I * for the 2π / 3 section (see FIG. 12A).

Next, in the step, ia ² + ic ² = I * ² for the section where the electrical angle is 0 to π / 3, and the generated torque during both-phase excitation τa (θe, ia, 0,0) + τa (θe + 2π / 3, The supply current ia, ic for each electrical angle θe, θe + 2π / 3 is calculated so that (0,0, ic) becomes maximum. Specifically, based on a plurality of actual measurement values stored in the pattern table (3),
As shown in FIGS. 12B and 12C, the supply current ia, ic is obtained.However, since the actual measurement values are only obtained discretely, for example, by performing interpolation calculation based on the actual measurement values. It is preferable to obtain a fine generated torque pattern and perform an operation based on the obtained generated torque pattern.

Then, in a step, by applying the obtained c-phase supply current pattern to the a-phase section [III],
As shown in FIG. 12D, it is possible to obtain the a-phase supply current pattern for the entire section.

FIG. 13 is a flowchart showing a method of calculating the supply current pattern of the a-phase (see FIG. 7C) based on the remarkably small measured value.
For the interval of 2π / 3, τ * 1 = τa (θe, ia, 0,0) (where τ * 1 is a preset relatively small predetermined value), and the supply current ia for each electrical angle θe is Ask. Specifically, based on a plurality of measured values, FIG.
As shown in, the supply current ia is obtained. However, since the generated torque pattern is only obtained discretely, for example, when the electrical angle is θe and the supply currents are ia1 and ia2, respectively. The generated torques are τa1 and τa2,
Moreover, if τa1 <τ * 1 <τa2, then the supply current ia
Is ia (θe) = ia1 + {(τ * 1-τa1) / (τa2τa
1)} (ia2-ia1). However, instead of the linear interpolation calculation as described above, a higher-order interpolation calculation may be performed to improve the approximation accuracy.

Next, in step, each electrical angle θe is such that τ * 1 = τa (θe, ia, 0,0) + τa (θe + 2π / 3,0,0, ic) for the section of electrical angle 0 to π / 3. , supply current ia, ic for θe + 2π / 3 is obtained. However, since there are many supply currents ia, ic that satisfy the above relationship, they cannot be uniquely defined. Therefore, specifically, for example, the pattern of the supply current ia in the section I is a linear pattern {ia = 0 when θ = θ0 as shown in FIG. 10B and ia = I1 when θ = θ1. (However, I1 is τ * 1 = τa (θ1, I1,0,0)
A current value that satisfies the above condition), and the supply current ia in section I and the supply current ic based on the above relation
(See FIG. 10C) may be calculated. And, in this case as well, it goes without saying that the interpolation calculation is performed as necessary.

Then, in step, supply current in section I
By adopting the pattern of ic as it is as the pattern of the supply current ia in the section III, as shown in FIG. 10D, the pattern of the supply current ia over the whole section of the electrical angle of 0 to π can be obtained.

Next, in a step, in the section where the electrical angle is π / 3 to 2π / 3, ia = I * 2 (I * 2 is a current command value corresponding to a relatively large torque command value τ * 2) Determine ia (see Figure 12A).

Then, in step, the electrical angle 0～Pai / for 3 sections ^{ia 2 + ic 2 = I *} 2 2 becomes, and the generated torque .tau.a during both phases energized (θe, ia, 0,0) + (τa (θe + 2π / 3,0,0, ic) is calculated so that the supply current ia, ic for each electrical angle θe, θe + 2π / 3 is maximized, specifically, multiple measured values stored in the pattern table (3). Based on the value
As shown in FIGS. 12B and 12C, the supply current ia, ic is obtained.However, since the actual measurement values are only obtained discretely, for example, by performing interpolation calculation based on the actual measurement values. It is preferable to obtain a fine generated torque pattern and perform an operation based on the obtained generated torque pattern.

Then, in a step, by applying the obtained c-phase supply current pattern to the a-phase section [III],
As shown in FIG. 12D, it is possible to obtain the a-phase supply current pattern for the entire section.

After the pattern of the supply current ia is obtained as described above, the pattern of the supply current ia can be made to correspond to the pattern of the supply current ib, ic by shifting the pattern of the supply current ia by 2π / 3. The three-phase reluctance motor (4) can be operated by controlling the supply current shown in the flowchart of FIG. 1 based only on the current pattern.

As is apparent from the above description, the pattern of the supply current to the three-phase reluctance motor (4) takes into consideration the influence of harmonics and magnetic saturation, and further, the phase corresponding to the rotation speed and the torque command value τ *. Since the advance angle is set, it is possible to greatly suppress the reduction of the generated torque due to the rotation speed and the torque command value, and to achieve the torque control close to the torque control determined based on the supply current pattern. .

Further, since it is only necessary to store the measured values discretely in the pattern table (2), it is possible to significantly reduce the number of necessary measured values, and also regarding the pattern of the supply current, the electrical angle 2π / It is sufficient to perform the calculation only within the range of 3, and the calculation amount can be significantly reduced.

Further, since the torque actually generated becomes linear with respect to the torque command value, the load on the host controller is reduced.

When the rotation direction of the three-phase reluctance motor (4) is set to be opposite, a series of processes similar to the above may be performed with the increasing direction of the electrical angle reversed.

Further, as is clear from the phase lead angle pattern shown in FIG. 5, when the torque command τ * is small, the phase shift angle is remarkably small and the generated torque can hardly be reduced. It is also possible to compare * with a predetermined threshold value and set the phase lead angle by the excitation timing changing unit (5) only when it is larger than the threshold value.
The deviation angle pattern is a quadratic curve as is apparent from FIG. 5, and therefore, the deviation angle pattern can be set by simulation instead of setting the deviation angle pattern based on the actual measurement value.

Note that the present invention is not limited to the above-described embodiment, and can be applied to, for example, torque control of a reluctance motor having an arbitrary number of phases, and a supply current pattern different from that of the above-described embodiment can be set. In addition to setting the supply current pattern shown in FIG. 7C, the magnitude of position deviation, speed deviation, etc. can be dealt with in accordance with the magnitude of FIG.
It is possible to selectively use the supply current pattern shown in A, B, further, it is possible to use a torque pattern instead of the supply current pattern, within the range not changing the gist of the present invention Various design changes can be made.

<Effects of the Invention> As described above, the first invention obtains the deviation angle of the excitation timing corresponding to the energization start time point and the energization end time point based on the rotation speed and the torque command value of the reluctance motor, and deviates them. Since the current is supplied to each phase at a timing shifted by an angle, the deviation between the supplied current and the actual winding current can be greatly suppressed, and the torque generated as the rotation speed and torque command value increase It is possible to suppress not only the decrease in the torque but also the generated torque, but it is possible to increase the generated torque.

According to the second aspect of the invention, it is possible to reduce the number of measured values of the deviation angle, and it is possible to greatly suppress the deviation between the supply current and the actual winding current by only performing a simple interpolation calculation. There is a unique effect that it is possible to greatly suppress the decrease in the generated torque due to the increase in the rotation speed and the torque command value.

According to the third aspect of the invention, the supply current waveform that greatly reduces the torque pulsation can be obtained based on a small number of measured values, and the obtained supply current waveform is supplied to each phase at the timing shifted by the deviation angle. The deviation between the supply current and the actual winding current can be greatly suppressed, the decrease in the generated torque due to the increase in the rotation speed and the torque command value can be greatly suppressed, and eventually the torque pulsation. It also has a unique effect that the increase of can be greatly suppressed.

According to the fourth aspect of the present invention, a supply current waveform that efficiently generates torque can be obtained based on a small number of measured values, and the obtained supply current waveform is supplied to each phase at a timing shifted by a deviation angle. The difference between the actual winding current and the actual winding current can be greatly suppressed, and the decrease in the generated torque due to the increase in the rotation speed and the torque command value can be greatly suppressed. Has the unique effect of being able to greatly suppress the decrease of

A fifth aspect of the present invention obtains a supply current waveform that greatly reduces torque pulsation and a supply current waveform that efficiently generates torque based on a small number of measured values, and selects one of the obtained supply current waveforms. Since the current is supplied to each phase at the timing shifted by the deviation angle, the deviation between the supply current and the actual winding current can be greatly suppressed. Therefore, it is possible to greatly suppress the decrease of the generated torque due to the increase of the rotation speed and the torque command value, and to greatly suppress the increase of the torque pulsation, and to reduce the generated torque over the entire range. It has a unique effect that it is possible to select a state in which can be greatly suppressed.

[Brief description of drawings]

FIG. 1 is a flow chart showing an embodiment of a torque control method of the present invention, FIG. 2 is a diagram for schematically explaining an excitation timing changing operation of a supply current pattern, and FIG. 3 is accompanying an excitation timing changing. FIG. 4 is a diagram schematically illustrating changes in winding current and generated torque for one phase, FIG. 4 is a diagram showing an example of a supply current pattern stored in a pattern table, and FIG. 5 is stored in an excitation timing changing unit. Showing an example of the phase lead angle pattern, FIG. 6 is a block diagram showing an example of the electrical configuration of the reluctance motor control device for implementing the torque control method of the present invention, and FIG. 7 is another example of the supply current pattern. FIG. 8, FIG. 11, FIG. 11 and FIG. 13 are flow charts for explaining the method for obtaining the supply current pattern, and FIG. 9 is the current supply state of each phase in the electrical angle range of 0 to 2π. Schematic diagram showing, Figure 10 is a schematic diagram illustrating a supply current pattern for sequentially calculating the torque pulsation, FIG. 12 is a sequential calculation, a supply current pattern for efficiently generating torque Schematic diagram for explaining, FIG. 14 is a diagram schematically showing a case where the phase shift angles on the energization start side and the energization end side are different, and FIG. 15 is a generated torque pattern corresponding to the case where both phase shift angles are equal and different. Fig. 16 is a diagram showing changes in self-inductance, Fig. 17 is a diagram showing a supply current pattern for flowing a current in a region where the inductance change is large, and Fig. 18 is a diagram showing supply current patterns in Fig. 17. The figure which shows the corresponding winding current and the generated torque for one phase. (1) …… Calculation unit, (2) …… Pattern table, (4) …… Three-phase reluctance motor, (4a) …… Rotor position sensor, (5) …… Excitation timing change unit, (6)… ... Position-speed converter, (? A) (? As) (? Ae) ... phase advance angle, [I] [II] [III] ... section

─────────────────────────────────────────────────── ─── Continuation of the front page (51) Int.Cl. ⁶ Identification code Office reference number FI technical display location H02P 8/00 303 A

Claims (5)

[Claims] 1. A torque control method for supplying currents having different electrical angles of 2π / n to each phase of an n-phase reluctance motor, wherein a current-carrying start time point and a current-carrying time are supplied based on a rotation speed and a torque command value of the reluctance motor. A torque control method for a reluctance motor, characterized in that a deviation angle of excitation timing is obtained corresponding to an end time point, and a current is supplied to each phase at a timing shifted by deviation angles (θas) (θae). 2. An excitation timing shift angle (θas) (θa
The torque control method for the reluctance motor according to claim 1, wherein e) is obtained by reading a deviation angle stored in advance in a two-dimensional table of the rotation speed and the torque command value and performing an interpolation calculation. . 3. The reluctance motor (4) is a three-phase reluctance motor, and the excitation interval for any one of the phases is three.
In the individual section [II], which is divided into individual sections [I], [II], and [III], and in which only the corresponding phase is excited, the generated torque is made constant based on the generated torque characteristics obtained from the measured data. A current pattern is calculated, an arbitrary supply current pattern is set for one of the complex sections [I] and [III] in which the corresponding phase and the other phase are simultaneously excited, and the other phase is arbitrarily set for the other of the complex sections. The supply current pattern that makes the generated torque constant when excited by the supply current pattern of is calculated based on the generated torque characteristics obtained from the measurement data, and the supply current pattern of each phase is calculated based on the continuous pattern. The reluctance according to claim 1 or 2, wherein the supply current is calculated and the calculated value of the current is supplied to each phase at a timing shifted by a shift angle (θas) (θae). Torque control method of chromatography data. 4. A reluctance motor (4) is a three-phase reluctance motor, and the excitation interval for any one of the phases is three.
The section [I], [II], [III] is divided into individual sections, and for the single section [II] in which only the relevant phase is excited, the supply current pattern that maximizes the generated torque is calculated based on the preset loss. However, for one of the complex sections [I] and [III] in which the corresponding phase and the other phase are simultaneously excited, the generated torque is maximized based on the generated torque characteristic obtained from the measured data and the preset loss. The supply current patterns of the relevant phase and other phases are calculated, and the composite section [I]
For the other side of [III], the supply current pattern of the other phase is applied, the supply current of each phase is calculated based on the pattern in which the supply current pattern is continuous, and the calculated value of the current is deviated by the deviation angle ( The torque control method for a reluctance motor according to claim 1 or 2, wherein the phases are supplied to each phase at a timing shifted by θas) (θae). 5. A reluctance motor (4) is a three-phase reluctance motor, and the excitation interval for any one of the phases is three.
In the individual section [II], which is divided into individual sections [I], [II], and [III], and in which only the corresponding phase is excited, the generated torque is made constant based on the generated torque characteristics obtained from the measured data. A current pattern is calculated, an arbitrary supply current pattern is set for one of the complex sections [I] and [III] in which the corresponding phase and the other phase are simultaneously excited, and the other phase is arbitrarily set for the other of the complex sections. A supply current pattern that makes the generated torque constant when excited by the supply current pattern is calculated based on the generated torque characteristics obtained from the measurement data, and a first pattern in which the supply current pattern is continuous is obtained in advance. At the same time, for the single section [II] in which only the relevant phase is excited, the supply current pattern that maximizes the generated torque is calculated based on the preset loss, and the relevant phase and other For one of the complex sections [I] and [III] in which is simultaneously excited, the supply current of the corresponding phase and other phases that maximize the generated torque based on the generated torque characteristics obtained from the measured data and the preset loss. A pattern is calculated, the supply current pattern of the other phase is applied to the other of the complex sections [I] and [III], and a second pattern in which the supply current pattern is continuous is obtained in advance to obtain the first pattern or The second pattern is selected, the supply current value of each phase is calculated based on the selected pattern, and the calculated value of the current is deviated by the deviation angle (θas) (θa
The torque control method of the reluctance motor according to claim 1 or 2, wherein the torque is supplied to each phase at a timing shifted by e).