JPH0574315B2 - - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Application of the Invention] The present invention uses a pulse width modulation (PWM) inverter to continuously control the current and direct a sinusoidal current to each phase winding depending on the rotor position. This is a speed control device for a brushless motor that uses a current flow method, and has a detector that specifically detects the position and speed of the rotor, and synchronizes the phase of the rotation speed and winding current based on these detection signals. The present invention relates to a speed control device for a brushless electric motor.

[Prior art]

Conventionally, resolvers have been used as speed and position detectors for plasticless motors, but these have disadvantages in that the detectors are larger than encoders and are more expensive. Further, when an absolute type encoder is used as the above-mentioned detector, there are disadvantages in that the resolution is generally lower than that of the incremental type, the electrical characteristics are poor, and the encoder cannot be used in a high-speed range.

In addition, when controlling the speed of a brushless motor using an incremental encoder, the absolute position of the rotor is not known when the control power is turned on, so the current command (reference) signal is not synchronized with the starting position detection signal, so the motor starts as is. If you do so, torque pulsation will occur.

[Purpose of the invention]

An object of the present invention is to provide an inexpensive speed control device for a brushless motor that suppresses a decrease in starting torque and has little torque pulsation.

[Summary of the invention]

The main points of the present invention will be explained using FIG.

Signal output from the distributor (magnetic pole position detector)
Based on PU, PV, and PW, signals (0, 0, 1), (1, 0, 1), etc. that give information on the position (angle) of the rotor in each 60° section, that is, the angle of the field magnetic flux. The angle data corresponding to these signals is stored in a memory, and the rotor position signal (θ _R ) is detected by capturing the change point of the rotor position detection code signal. For example, in FIG. 1, the point in time when the detection code changes from (0, 0, 1) to (1, 0, 1) is captured, position data is read from the corresponding memory, and 0° is given as the rotor position θ _R. Also,
Detection code is from (1, 0, 1) to (1, 0, 0)
When it changes to 60° is given as θ _R.

Current command signal (α) based on the above θ _R signal
and the rotor position (field magnetic flux) are synchronized so that the phase difference between them is zero. Synchronization at startup when the control power is turned on is performed as follows. First, the position detection code for the 60° interval (1, 0, 1), (1, 0, 0),...
Read (0, 0, 1), rotor position (angle)
is 0° to 60°, 60° to 120°, 300° in 0° to 360°
~
Detect which 60° section of 360° it is in, and determine the center value of the corresponding section (for example, if it is found to be in the -60° to 0° section in Figure 1), set it as the initial phase of the current command signal. -30°).

When the rotor is driven and the rotor detection code first changes, the rotor position θ _R setting changes from (0, 0, 1) to (1, 0, 1) in Figure 1. Set 0° as the time θ _R , and find the deviation Δθ′ between the phase θ ₁ of the current command signal and the above θ _2. This deviation gradually becomes zero every time the speed control system is started. Perform automatic corrections.

Next, let's evaluate the maximum torque fluctuation at the initial start when the power is turned on. Assuming that the field magnetic flux has a sinusoidal distribution, the instantaneous generated torque τ _M is given by the following equation.

τ _M K _M {I ₁ sin(θ _R +Δθ′)φsinθ _R +I ₁ sin(
θ _R −120°+Δθ′×φsin(θ _R −120°) +I ₁ sin(θ _R −240°+Δθ′)φsin(θ _R −2
40°)}=3/2K _M・φ・I ₁ cosΔθ′……(1) However, K _M : Proportionality constant φ: Effective field magnetic flux I ₁ : Amplitude value of each winding current of U, V, W (1) From the formula, the maximum torque fluctuation caused by the phase difference between the current command signal and the rotor at startup is (1-
cos Δθ') and set (Δθ') = 30°. This value is about 13%, and this variation is compensated by the speed control system until the next 60° section detection signal is generated.

Even in normal operating conditions, the phase θ ₁ of the current command signal is corrected by referring to the rotor position θ _R obtained at every 60° timing, and the error accumulated in the phase θ ₁ of the current command signal is corrected. Very little. If such a method is not used, if there is an error of ±Δω ₁ in the angular frequency ω ₁ of the current command signal, the phase θ ₁ will be ∫
Since it is always integrated in the form of (ω ₁ ±Δω ₁ ) dt, if it is operated vertically for a long time, errors will increase and synchronization will occur. Therefore, it is necessary to constantly correct the value of θ ₁ even in normal operating conditions.

[Embodiments of the invention]

FIG. 2 shows a control block configuration, and FIG. 3 shows a system configuration for driving a brushless motor at variable speed using the control system shown in FIG.

Analog from speed setting of potentiometer etc.
The speed command ω REF is converted into a digital quantity, the speed command ω _REF is taken in, the number N of encoder pulses within the sampling time obtained from the speed measurement circuit, and the time width T of the signal obtained by synchronizing the signal with the sampling period and the encoder signal. The rotational angular velocity ω _M obtained from the velocity calculation process 21 is determined based on each data. The deviation Δε between ω _REF and ω _M is determined, speed calculation compensation (proportional + integral compensation) 20 is performed so that Δε becomes zero, and the amplitude value I ₁ of the current command signal is determined and stored in the amplitude value storage register 24. Torque control is performed by changing the value of _I1 . On the other hand, the rotational angular velocity ω _M obtained by the speed calculation process 21
The angular frequency (inverter angular frequency) ω ₁ of the current command signal is once obtained through gain conversion processing 22 and stored in the angular frequency storage register.

The above ω ₁ is determined by the integrator 25 as an angle (phase)
Obtain θ ₁ . Synchronization of the rotor position and the current command signal is performed as follows. Position detector (distributor)
Processing 2 for calculating data for synchronization corresponding to the change point of each 60° interval signal based on information obtained from the 60° interval detection signal forming circuit 29
7, find θ _R , find the deviation Δθ between this and θ ₁ obtained by reading the contents of the integrator 25, and convert this to 1
Correction data Δθ' for synchronization is obtained by the next lag compensation 26, and the above data Δθ' is added to θ ₁ to obtain the phase θ ^* of the current command signal synchronized with the rotor position, and θ ^* and (θ ^* −120°) are obtained from the sine wave storage memory 28, and these values are stored in the D/
Two-phase analog current command with A converter 30
Find I ₁ sin θ ^* and I ₁ sin (θ ^* −120°), and then
Based on the phase, the 2-phase to 3-phase conversion circuit 31 generates phase current command signals I ₁ sin θ ^* , I ₁ sin (θ ^* −120°), I ₁ sin (θ
^* −
240°). The three-phase current command signals mentioned above are compared with the corresponding winding phase currents by a comparator 32 to obtain a pulse width modulation (PWM) signal.

FIG. 3 shows the configuration of a brushless variable speed drive system using the above-mentioned algorithm.

AC power supply 1 is rectified by rectifier 2 to obtain DC voltage, which is smoothed by smoothing capacitor 3 and PWM
Input to inverter 4. The PWM inverter applies variable voltage/variable frequency voltage to the brushless motor 6 to drive it at variable speed.

The above PWM inverter is controlled by the following configuration. A rotor position detector 7 and an encoder 8 are attached to the electric motor 6, and each
60° section detection signal formation circuit 29, speed measurement circuit 1
4 is input. From the circuits 29 and 14, the second
Interrupt signals IRQ ₂ and IRQ ₁ are generated to start the programs of synchronization control processing 26, 27 and speed control processing 20, 21, 22 shown in the figure.

When the interrupt signal IRQ ₁ is generated, the microprocessor (MPU) 9 specifies the EPROM 12 using the address signal generated from the address decoder 10, and executes the speed control processing and phase correction data calculation processing programs stored therein. Execute. The resulting data I ₁ , ω ₁ , Δθ' are stored in a predetermined register of the current command circuit 15. At 15
Based on the data of I ₁ , ω ₁ , Δθ′, three-phase current command signals I ₁ sin (∫ω ₁ dt + Δθ′), I ₁ sin (∫ω ₁ dt−120°
+Δθ′),
I ₁ sin (∫ω ₁ dt−240°+Δθ′) and the above 3
The phase command signals are input to the current control circuit 16 and compared with each phase winding current obtained from the current detector 5.
Get PWM signal. This PWM signal is amplified by the base drive circuit 17 and applied to the base of the PWM inverter 4.

Furthermore, when the interrupt signal IRQ ₂ is generated, a synchronization control processing program is executed to match the phase of the rotor position and the current command signal. First, the state of the circuit 29 is read and the corresponding rotor phase θ _R is determined. Next, read the data θ ₁ = (∫ω ₁ dt) from the current command signal phase storage register of the current command circuit, find the deviation Δθ (= θ _R − θ ₁ ) of the above θ _R , and perform first-order lag compensation. Then, phase correction data Δθ' is obtained and stored in the phase correction storage register of the current command circuit 15.

The circuit 15 adds Δθ' to the above θ ₁ to obtain the synchronized phase θ ^* , which is used as the phase command of the actual current command signal. A PWM signal based on this command is generated in circuit 16, and this signal is amplified in circuit 17 to become a base signal.

Note that this will be executed between the current IRQ ₂ and the next IRQ ₂ .
The phase correction data calculation processing program in IRQ ₁ is based on Δθ′ obtained as a result of the current IRQ ₂ processing.
Compensate for the next delay.

That is, in the processing of IRQ ₂ , Δθ=θ _R −θ ₁ ...(2) (Δθ') = Δθ/1+TS ...(3) T: time constant S: differential operator, and in the processing of IRQ ₁ , (Δθ ′) _o = Δθ′ _o-1 /1+TS ...(4).

However, (Δθ′) _o-1 : Value obtained by the first-order lag compensation process for the previous IRQ ₁ (Δθ′) _o : Value obtained by the first-order lag compensation process for the current IRQ ₁ Next, interrupt signals IRQ ₁ in circuits 14 and 29,
We will explain how IRQ ₂ occurs.

FIG. 4 shows a timing chart of interrupt signal generation. A signal Ec is output from the encoder, and this is synchronized with a signal T _S0 having an activation period for executing a program driven by an interrupt signal IRQ ₁ output from the timer of the circuit 14 to obtain a signal T _S1 . . Edge of signal T _S1 (rising,
(at the falling edge) to the interrupt signal IRQ ₁ shown in Figure 4.
get.

In addition, the signals PU, PV, and PW obtained from the rotor position detector (distributor) 7 are synchronized with the encoder signal E _c to obtain PU', PV', and PW', and from the edges of these three signals, as shown in Figure 4. Obtain the interrupt signal IRQ ₂ shown in .
The reason for synchronizing with the encoder signal is IRQ ₁ ,
This is to improve the accuracy of speed measurement in each IRQ ₂ process.

The interrupt signals that enter the MPU are the above IRQ ₁ and
Let IRQ be the signal obtained by taking the NOR with IRQ ₂ . Note that the priority of the interrupt signals IRQ ₁ and IRQ ₂ is such that the interrupt signal of IRQ ₂ , which performs synchronization processing, is given a higher priority.
This is done because it is necessary to perform synchronous data processing with as little time delay as possible.

Next, FIG. 5 shows a specific configuration of the current command circuit 15. FIG. 6 shows a time chart explaining the operation.

When the operation command a is input, it is converted into a signal b (synchronized at the falling edge in the figure) which is synchronized with the enable signal E output from the MPU, and further a signal c which is synchronized with the signal E is obtained. An initial activation signal STC IN is obtained by taking the EOR of the signals b and c to form stage signals STC0 to STC11. Operation command a above
A stage signal generation circuit 56 generates a signal for the STC 11 from the signal. These 56 circuits can be easily constructed using latches, shift registers, and the like. 5
Based on the signal generated at step 6, the hardware processing shown in the control block of FIG. 2 is executed.

Here, the data ω ₁ , I ₁ , and Δθ′ (absolute value and sign data) obtained by the interrupt routines IRQ ₁ and IRQ ₂ are temporarily stored in the PIA 50 by the MPU via the data bus. (here, ω ₁ , I ₁ , Δθ′: 2-byte data). The absolute value of each of the data ω ₁ and Δθ' is stored in 53 dedicated registers, and the code data is stored in register 52. The strobe signals for which ω ₁ and Δθ' are transferred to the above registers are shown in Figure 6 FRE Reg ST, PHA Reg
Indicated by ST. These strobe signals come from the MPU
While transferring to PIA50, strobe signals as shown by broken lines _P16 and _P15 are suppressed (while this signal is generated, the corresponding FRE Reg ST,
Suppress each signal of PHA Reg ST. ). This is because the 2-byte data is stored in the register 53 in an aligned manner.

The current amplitude value (signed data) is transferred to the register of the D/A converter 54, and 2-byte data is aligned and D/A converted.

The data of ω ₁ is the angle θ _o-1 (=
(ω ₁ ′∫dt) _o-1 ) is the addition (subtraction) stored in 52.
Addition (subtraction) is performed by the acid command (ω ₁ ±θ _o-1
→θ _o ), and the resulting data is sent to the register 5 by the A signal of the strobe signal TEM Reg ST.
9, and further stored in register 57 by ADD Reg ST.

Next, the phase correction data Δθ is outputted from a predetermined register 53 by the output control signal PHA Reg OUT, and is added (subtracted) to the content θ _o of the register 57 by the addition (subtraction) command 52. The result is stored in register 59 by TEM Reg ST. The resulting data θ ^* (=θ _o ±Δθ′) is sent to the D/A converter 30 by the strobe signal DAST.
A current command signal whose amplitude is modulated by the output value of 54 is generated. 30, θ ^* , θ ^* −
A two-phase current command signal corresponding to 120° is generated, which is converted into two-phase to three-phase in circuit 31, and this and the corresponding primary current are compared in circuit 32.
Get PWM signal.

In the above example, the speed measurement process is performed in the IRQ ₁ routine, but in order to compensate for the phase shift as accurately as possible during sudden acceleration/deceleration, the speed measurement process is also performed in the IRQ ₂ routine every 60° interval. Just do it like this.

〔Effect of the invention〕

According to the present invention, the rotating magnetic pole position and the winding current phase can be synchronized using an inexpensive distributor without using an expensive resolver or absolute encoder. This has the effect of realizing an inexpensive speed control device for a brushless motor with less pulsation.

[Brief explanation of the drawing]

FIG. 1 is a diagram explaining the principle of synchronization of the rotor phase and current command signal according to the present invention, FIG. 2 is a control block diagram of the brushless motor according to the present invention, and FIG. 3 is a system configuration diagram according to the present invention. 4 is an interrupt signal generation time chart, FIG. 5 is a current command circuit diagram, and FIG. 6 is a time chart explaining the operation of FIG. 5. 20...Speed calculation processing, 21...Speed measurement processing, 25...Integrator, 26...First-order delay compensation element, 27...Synchronization data calculation processing, 28...Sine wave table memory, 29...60 °Section detection signal forming circuit, 24, 30...D/A converter, 31...
...2-phase to three-phase conversion circuit, 32... Comparator, 4...
PWM inverter, 5...Current detector, 6...Brushless motor, 7...Distributor (rotor position detector), 8...Encoder, 9...Microprocessor, 11, 12...Memory, 14... ... Speed measurement circuit, 29 ... 60° position detection circuit, 15 ... Current command circuit, 16 ... Current control circuit, 17 ... Base drive circuit.

Claims (1)

[Scope of Claims] 1. A speed control device for a brushless motor that controls the magnitude and phase of a winding current of a three-phase brushless motor having rotor magnetic poles and drives the motor at variable speed using a pulse width modulation inverter, comprising: means for commanding a speed; a speed detector for detecting the rotational speed of the rotor; means for creating a command signal for the magnitude of the winding current from the rotational speed detection signal and the rotational speed command signal; means for creating the winding current phase command signal from a rotational speed detection signal; a distributor for outputting a three-phase square wave signal having a phase difference of 120 degrees according to the rotational position of the rotor of the motor; means for outputting a position detection code that changes every 60 degrees of the rotor position from the three-phase square wave signal; means for detecting a change in the state of the rotor position detection code every 60 degrees; 0 of the rotor corresponding to the detection code
storage means for storing position signal data (equivalent to 0, 60, ... 300, 360 degrees) for each 60 degree section within a position of 360 degrees from 360 degrees; and inputting information at the time of change of the detection code into the storage means. The position signal data corresponding to the information is retrieved, the deviation between the position signal data and the signal obtained by integrating the rotational speed signal is calculated, and the value obtained by first-order lag compensation of the deviation value is calculated as the value obtained by the winding current. means for correcting a phase command signal; a pulse width modulation inverter that drives the brushless motor at a variable speed by inputting a winding current phase command signal obtained by the correcting means and a command signal for the magnitude of the winding current; A speed control device for a brushless motor, characterized by comprising: