JP6252843B2 - Magnetic pole position adjusting method and magnetic pole position adjusting apparatus for synchronous motor - Google Patents

Description

  The present invention relates to a magnetic pole position adjusting method and a magnetic pole position adjusting apparatus for performing so-called magnetic pole positioning that eliminates an error between a magnetic pole position detection value of a synchronous motor and an actual magnetic pole position.

  In a synchronous motor equipped with a position detector such as a resolver, a current of a predetermined phase is passed from a power converter such as an inverter to a stator winding of the synchronous motor based on a magnetic pole position detection value output from the position detector. To generate a rotating magnetic field. Here, since the normal position detector detects only the relative position of the magnetic pole, the initial value of the magnetic pole position is adjusted by magnetic pole alignment in order to accurately detect the magnetic pole position and control the synchronous motor. There is a need.

As a conventional magnetic pole positioning method, the stator winding of the motor is DC-excited to fix the rotor position, and the rotor position is set to the initial value (the magnetic pole position is 0 °). A method for adjusting the detection value is known.
However, in this method, there is a possibility that the rotor does not stop at a fixed position due to static friction or the like, and as a result, there is a problem that the magnetic pole alignment cannot be performed accurately.

For this reason, Patent Documents 1, 2 and the like have proposed a method of performing magnetic pole alignment without relying on a direct DC excitation of the electric motor.
For example, in Patent Document 1, as shown in FIG. 4, the error power ΔP between the DC input power calculation value P _DC of the inverter 11 and the machine output calculation value P _m of the synchronous motor SM is a value corresponding to the magnetic pole position error. The correction amount Δθ _f based on the error power ΔP, the rotational speed n, etc. is added to the magnetic pole position detection value θ _f for correction, thereby calculating the true magnetic pole position θ _f ′ and torque control. Accuracy is improved. In FIG. 4, 12 is a DC power source, 13 is a speed detector, 14 is a position detector, and 15 is a window comparator.

Patent Document 2 discloses that magnetic pole alignment is performed by the processes shown in FIGS. 5 and 6.
In the example of FIG. 5, a correction angle for adjusting the relative deviation between the magnetic pole position detection value and the actual magnetic pole position is temporarily set in advance (step S1), and speed feedback obtained by differentiating the magnetic pole position detection value. Is taken into the control device and it is determined whether or not it is equal to or greater than the overspeed set value (S2, S3). While this speed feedback is less than the overspeed set value (No in S3), the process (S4, S5) of executing speed control according to a predetermined speed reference is repeated while gradually shifting the correction angle in step S1. When the speed feedback becomes equal to or higher than the overspeed set value, the motor is set in a free-run state, and the temporarily set correction angle is temporarily stored (S3 Yes, S6). Next, a series of processing is repeated while shifting the temporarily set correction angle by a predetermined value in the reverse direction (S7 No, S1 to S5), and the correction angle when the speed feedback becomes equal to or higher than the overspeed set value is again set. Store (S3 Yes, S6).
An intermediate point between the two correction angles thus obtained is calculated to obtain a final correction angle (S7 Yes, S8), and the magnetic pole position is adjusted using this correction angle.

  Further, in the example shown in FIG. 6, speed control is performed according to the speed reference generated under the temporarily set correction angle (steps S <b> 11 to S <b> 14), and an evaluation function for calculating the speed control characteristic is calculated. The process is repeated a predetermined number of times while gradually shifting the correction angle (S15, S16). Thereafter, a correction angle is determined so that the evaluation value of the speed followability with respect to the speed reference is the best (S17), and the magnetic pole position is adjusted using this correction angle.

JP-A-8-308292 (paragraphs [0017] to [0022], FIG. 3 etc.) JP 2009-72033 A (paragraphs [0014] to [0019], [0021] to [0027], FIG. 2, FIG. 5, etc.)

However, the conventional technique according to Patent Document 1 requires arithmetic processing and a circuit for obtaining the error power ΔP, and the conventional technique according to Patent Document 2 performs a process of performing speed control while shifting the correction angle. The correction angle had to be performed in both forward and reverse directions, and much time and labor were required for magnetic pole alignment.
SUMMARY OF THE INVENTION An object of the present invention is to provide a magnetic pole position adjustment method and a magnetic pole position adjustment apparatus for a synchronous motor that can perform magnetic pole position alignment by a simple calculation.

In order to solve the above-described problem, a magnetic pole position adjustment method according to claim 1 corrects the magnetic pole position detection value obtained by the position detector of the synchronous motor with an offset value, so that the magnetic pole position detection value and the actual magnetic pole position are In the magnetic pole position adjustment method for performing the control calculation of the synchronous motor by calculating the magnetic pole position without the error of
The stator winding of the synchronous motor is DC-excited to detect a plurality of first inflection points at which the change rate of the magnetic pole position detection value changes from positive to negative, and the change rate changes from negative to positive. A plurality of second inflection points are detected, and a first envelope passing through the plurality of first inflection points and a second envelope passing through the plurality of second inflection points are respectively obtained by linear approximation,
Times of intersections of the first envelope and the second envelope are obtained by detecting the magnetic pole position detection values respectively corresponding to the plurality of first inflection points and the second inflection points, the first envelope, and the Calculated from the slope of the second envelope ,
The offset value corresponding to the intersection is calculated by substituting the calculated time of the intersection into the first envelope or the second envelope .

The magnetic pole position adjusting device according to claim 2 corrects the magnetic pole position detection value by the position detector of the synchronous motor with an offset value, thereby eliminating an error between the magnetic pole position detection value and the actual magnetic pole position. In the magnetic pole position adjusting device for calculating the synchronous motor and calculating the synchronous motor,
A direct current excitation unit for direct current excitation of the stator winding of the synchronous motor;
A first envelope passing through a plurality of first inflection points where the change rate of the magnetic pole position detection value changes from positive to negative, and a first envelope passing through a plurality of second inflection points where the change rate changes from negative to positive. An envelope calculation unit for calculating two envelopes by linear approximation,
Times of intersections of the first envelope and the second envelope are obtained by detecting the magnetic pole position detection values respectively corresponding to the plurality of first inflection points and the second inflection points, the first envelope, and the and an offset calculating unit for calculating the offset value by the second calculated from the slope of the envelope, the calculated the intersection of time are substituted into the first envelope and said second envelope corresponding to said intersection, It is equipped with.

According to the present invention, magnetic pole alignment can be performed with high accuracy without being affected by static friction or the like, compared to a method of obtaining an initial value from a position where a rotor is stopped by direct current excitation of a synchronous motor.
In addition, there is no need for arithmetic processing or a circuit for obtaining error power as in Patent Document 1, and speed control is performed in both the forward and reverse directions of the correction angle as in Patent Document 2, for example, by shifting the correction angle. Since complicated processing is not required, cost associated with magnetic pole alignment can be reduced, speeded up, and simplified.

It is a block diagram of the drive system of the synchronous motor to which the embodiment of the present invention is applied. It is a block diagram of the initial value calculating part in FIG. It is a wave form diagram which shows operation | movement of the initial value calculating part in FIG. It is a block diagram of the prior art described in patent document 1. FIG. 10 is a flowchart showing the operation of the prior art described in Patent Document 2. 10 is a flowchart showing the operation of the prior art described in Patent Document 2.

Hereinafter, embodiments of the present invention will be described with reference to the drawings.
FIG. 1 is a configuration diagram of a drive system for a synchronous motor to which an embodiment of the present invention is applied.
In FIG. 1, a DC power source 1 is connected to a three-phase inverter 2, and a synchronous motor 4 is connected to the output side via a current detector 3. A resolver 5 as a position detector for detecting a magnetic pole position (angle) is attached to the rotor of the synchronous motor 4, and an output signal thereof is input to the control device 6. Reference numeral 7 denotes a load driven by the synchronous motor 4.

Next, the configuration and operation of the control device 6 will be described.
In the control device 6, the operation command, the rotational speed command ω ^* and the rotational speed detection value ω are input to the torque command unit 61, and the torque command T ^* generated based on these is input to the inverter control unit 62. Has been.
On the other hand, the magnetic pole position detection value θ ₀ calculated by the position detector 63 based on the output signal of the resolver 5 is added to the initial magnetic pole position θ ₁ by the adder 64 and the added value θ ₂ (= θ ₁ + θ ₀ ) is output. Note that θ ₀ , θ ₁ , and θ ₂ are all electrical angles. Here, the initial magnetic pole position θ ₁ (offset value θ _offset ) corresponds to an error between the magnetic pole position detection value θ ₀ by the resolver 5 and the actual magnetic pole position, and is obtained by the initial value calculation unit 66.

In this embodiment, the initial value calculation unit 66 obtains an offset value θ _offset by the method shown in FIGS. 2 and 3 to be described later, and the magnetic pole position detection value θ ₀ is corrected by the offset value θ _offset to control the magnetic pole in the control calculation. The position θ ₂ is generated and used for torque control or the like.

Adder 64 pole position theta ₂ which is output from the rotational speed detection value is differentiated ω is calculated by differentiating circuit 65, the speed detection value ω is input to the torque command unit 61, as described above. Moreover, the speed detection value omega, the torque command T ^*, is input to the inverter control section 62 together with the output current and the magnetic pole position theta ₂ of the inverter 2. The inverter control unit 62 generates and outputs a gate signal for turning on and off the semiconductor switching element (not shown) of the inverter 2 based on the input information.

Next, FIG. 2 shows a configuration of the initial value calculation unit 66, which is realized by software of a microcomputer, for example.
In FIG. 2, the direct current excitation unit 66 a activates the inverter 2 via the inverter control unit 62 based on the magnetic pole alignment command, and direct current excites the stator winding of the synchronous motor 4.
Position and time detector 66b from the magnetic pole position detection value theta ₀ by the position detection unit 63 of FIG. 1, the negative rate of change is positive, or, the value when the inverted (so-called inflection point) from negative to positive _{θ 11, θ 21, θ 12} , θ 22, detected along the ... the time axis, the time _t 11 corresponding to _{each, t 21, t 12, t} 22, stored with ....... The envelope calculation unit 66c uses the magnetic pole position detection values θ ₁₁ , θ ₂₁ , θ ₁₂ , θ ₂₂ ,... And the times t ₁₁ , t ₂₁ , t ₁₂ , t ₂₂ ,. One of the calculated envelope, offset calculating section 66d calculates the offset value theta _offset from the intersection of the two envelope the envelope calculation section 66c is determined.

FIG. 3 is a waveform diagram showing the operation of the initial value calculator 66 described above.
Conventionally, when the synchronous motor 4 is DC-excited to change the magnetic pole position detection value θ ₀ as shown in FIG. 3 and the rotor stops at time t _stop , the magnetic pole position detection value θ _stop at that time is set to the initial value. Although the output of the position detector is corrected as described above, there is no guarantee that this detected value θ _stop will always be a constant value due to static friction or the like. For this reason, the reliability of the initial value is enhanced by a method of obtaining an average value by performing an operation for obtaining θ _stop a plurality of times. However, according to this method, much time is required for magnetic pole alignment.
Therefore, in this embodiment, the offset value θ _offset is obtained by the following method to determine the initial magnetic pole position.

That is, in FIG. 3, the magnetic pole alignment command at time t ₀₀ is input to the DC excitation portion 66a of FIG. 2, the magnetic pole position detection value from the position detecting section 63 the rotor is rotated by the DC excitation of a synchronous motor 4 theta ₀ Is output. Position and time detector 66b has a value in the first inflection point at which the polarity of the rate of change of the magnetic pole position detection value theta ₀ changes and theta _11, stores the time corresponding thereto as t _11. Here, the time _{t 11,} put a time _{t 0} of the base of the envelope _f 1, _{f 2,} which will be described later.
Thereafter, similarly, the magnetic pole position detection values θ ₂₁ , θ ₁₂ , θ ₂₂ ,... At the inflection points and the times t ₂₁ , t ₁₂ , t ₂₂ ,. I will do it. Here, data θ ₁₁ , t ₁₁ , θ ₁₂ , t ₁₂ when the change rate of the magnetic pole position detection value θ ₀ is inverted from positive to negative, and data θ ₂₁ , t ₂₁ , when the change rate from negative to positive is inverted. The envelopes f ₁ and f ₂ are determined using θ ₂₂ and t ₂₂ .

Next, the envelope calculation unit 66c uses the two sets of data θ ₁₁ , t ₁₁ , θ ₁₂ , t ₁₂ and θ ₂₁ , t ₂₁ , θ ₂₂ , t ₂₂ to perform the envelope f by linear approximation. _1, determine the _{f 2.}
Here, the magnetic pole position detection values θ ₁₁ , θ ₁₂ ,... Are the first inflection point, the magnetic pole position detection values θ ₂₁ , θ ₂₂ ,... Are the second inflection point, and the envelope f ₁ is first envelope, the envelope _{f 2} to the second envelope, corresponding respectively.

As apparent from FIG. 3, the inclination _{k 1} of the envelope _{f 1} by Equation 1, the slope _{k 2} of the envelope _{f 2} by Equation 2, obtained respectively.
[Formula 1]
(Θ ₁₂ −θ ₁₁ ) / (t ₁₂ −t ₁₁ ) = k ₁
[Formula 2]
(Θ ₂₂ −θ ₂₁ ) / (t ₂₂ −t ₂₁ ) = k ₂

In addition, the offset values of the envelopes f ₁ and f _{2 at} the starting time t ₀ are set as Formula 3 and Formula 4, respectively.
[Formula 3]
θ ₁₀ = θ ₁₁
[Formula 4]
θ ₂₀ = θ ₂₂ −k ₂ × (t ₂₁ −t ₀ )

As a result, the envelope f ₁ can be expressed by Equation 5 and the envelope f ₂ can be expressed by Equation 6.
[Formula 5]
f ₁ = k ₁ × t + θ ₁₀
[Formula 6]
f ₂ = k ₂ × t + θ ₂₀
Therefore, if f ₁ = f _{2 is established} , the time t (t _x in FIG. 3) when the following formula 7 is established is obtained by the formula 8, and this time t _x is substituted into the formula 5 or the formula 6, then the envelope The detected magnetic pole position value θ ₀ at the intersection of f ₁ and f ₂ can be calculated as the offset value θ _offset .
[Formula 7]
k ₁ × t + θ ₁₀ = k ₂ × t + θ ₂₀
[Formula 8]
t = (θ ₂₀ −θ ₁₀ ) / (k ₁ −k ₂ )

The envelope calculation unit 66c in the initial value calculation unit 66 of FIG. 2 calculates the envelopes f ₁ and f ₂ using Formulas 1 to 6, and the offset calculation unit 66d calculates the time t _x using Formulas 7 and 8 and Formula 5 _{Alternatively} , the offset value θ _offset is calculated by Equation 6. By adding the offset value θ _offset to the magnetic pole position detection value θ ₀ by the adder 64, the magnetic pole position θ ₂ in the control calculation in which the magnetic pole position error is corrected can be obtained.

That is, when the magnetic pole position detection value θ _stop at the time t _{stop when the} rotor is stopped by direct current excitation is regarded as an initial value as in the prior art, an error occurs in the magnetic pole position detection value θ _{stop due} to static friction or the like. However, according to this embodiment, the offset value θ _offset is uniquely defined based on the intersection points of the first and second envelopes obtained from the plurality of first and second inflection points that are not affected by static friction or the like. Therefore, magnetic pole alignment can be performed with high accuracy.

Note that when three or more points of inflection point data as an envelope, for example, _{θ 11, t 11, θ 21} , t 21, θ 12, t 12 is obtained, with an envelope according to a quadratic function approximation offset The value θ _offset may be obtained. In this case, the time t _x for calculating the offset value θ _offset can be obtained from the solution {−b ± √ (b ² −4ac) / 2} of the quadratic equation. Alternatively, the envelopes f ₁ and f ₂ may be obtained by the least square method, and the time t _x and the offset value θ _offset may be obtained.

1: DC power supply 2: Inverter 3: Current detector 4: Synchronous motor 5: Resolver (position detector)
6: Controller 61: Torque command section 62: Inverter control section 63: Position detection section 64: Adder 65: Differentiation circuit 66: Initial value calculation section 66a: DC excitation section 66b: Position / time detection section 66c: Envelope calculation Unit 66d: offset calculation unit 7: load

Claims (2)

By correcting the magnetic pole position detection value by the position detector of the synchronous motor with the offset value, the magnetic pole position that eliminates the error between the magnetic pole position detection value and the actual magnetic pole position is calculated, and the control calculation of the synchronous motor is performed. In the magnetic pole position adjustment method for performing,
The stator winding of the synchronous motor is DC-excited to detect a plurality of first inflection points at which the change rate of the magnetic pole position detection value changes from positive to negative, and the change rate changes from negative to positive. A plurality of second inflection points are detected, and a first envelope passing through the plurality of first inflection points and a second envelope passing through the plurality of second inflection points are respectively obtained by linear approximation,
Times of intersections of the first envelope and the second envelope are obtained by detecting the magnetic pole position detection values respectively corresponding to the plurality of first inflection points and the second inflection points, the first envelope, and the Calculated from the slope of the second envelope ,
A method for adjusting a magnetic pole position of a synchronous motor, wherein the offset value corresponding to the intersection is calculated by substituting the calculated time of the intersection into the first envelope or the second envelope . By correcting the magnetic pole position detection value by the position detector of the synchronous motor with the offset value, the magnetic pole position that eliminates the error between the magnetic pole position detection value and the actual magnetic pole position is calculated, and the control calculation of the synchronous motor is performed. In the magnetic pole position adjusting device for performing
A direct current excitation unit for direct current excitation of the stator winding of the synchronous motor;
A first envelope passing through a plurality of first inflection points where the change rate of the magnetic pole position detection value changes from positive to negative, and a first envelope passing through a plurality of second inflection points where the change rate changes from negative to positive. An envelope calculation unit for calculating two envelopes by linear approximation,
Times of intersections of the first envelope and the second envelope are obtained by detecting the magnetic pole position detection values respectively corresponding to the plurality of first inflection points and the second inflection points, the first envelope, and the and an offset calculating unit for calculating the offset value by the second calculated from the slope of the envelope, the calculated the intersection of time are substituted into the first envelope and said second envelope corresponding to said intersection,
A magnetic pole position adjusting device for a synchronous motor, comprising:

Images