CN103414427A - Brushless direct current motor control method - Google Patents

Abstract

The invention provides a brushless direct current motor control method. According to the method, only through the utilization of a simple switch-type Hall position detector and an observer obtained through software, relatively accurate observation values of the motor turn angle and the motor speed are obtained through reconstruction, and the sine wave vector control over a brushless direct current motor is achieved though the observation values. The method not only keeps the advantages of being simple, reliable and low in cost of a brushless motor control system, but also improves control accuracy and dynamic performance.

Description

Control method for brushless direct current motor

Technical field

The present invention relates to a kind of control method of brushless DC motor, particularly a kind of sine wave vector of brushless DC motor based on observer control method.

Background technology

According to the difference that the phase excitation magnetic potential distributes, three-phase permanent-magnetic synchronous motors can be divided into two types: sinusoidal waveform and square wave type permanent magnet synchronous motor, and it is sinusoidal wave shape that the former excitation magnetic potential distributes, latter's excitation magnetic potential distributes and is square-wave-shaped.

The square wave type permanent magnet synchronous motor also is called brushless DC motor (BLDC); Sinusoidal waveform permanent magnet synchronous motor custom is called as PMSM(Permanent Magnet Synchronous Motor).According to traditional technology, sine-wave permanent magnet synchronous motor (PMSM), generally adopt the sine wave vector control mode, for high-precision position servo control, needs more high-precision position detector, as photoelectric pulse coder, resolver etc.Brshless DC motor (BLDC) generally adopts the square wave control mode, only for simple speed, controls, and easy Hall switch position detector only is housed, and its direct position detection accuracy only has 60 degree (electrical degree).The main feature of brushless DC motor control system is that simple and reliable cost is low, and shortcoming is that control precision and dynamic property are lower.

Brshless DC motor sine wave vector control method is in research always in recent years, the main problem that needs to solve is in the situation that only adopt simple hall detector, improve position detection accuracy as far as possible, common solution is: suppose that the previous hall signal cycle is identical with the current hall signal cycle, pass through Measuring Time, indirectly calculate the motor corner, this method in speed low and motor to be under the velocity variations state be inaccurate, thereby also limited its practicality.

Summary of the invention

Technical problem to be solved by this invention: there is defect in the present invention in order to solve in prior art, the corner of brshless DC motor, speed and load observation model and control method thereof, the method is only utilized easy Hall switch position detector, realizes that the sine wave of brshless DC motor is controlled.

Technical scheme of the present invention is:

Control method for brushless direct current motor comprises the following steps:

Step 1, set up corner, speed and the load observation model of brshless DC motor, specific as follows:

(1) the sense cycle T to fix, read the three road position pulse signals that the motor hall detector is sent in real time, and the real-time total number of Ji Qu tri-road P-pulse obtains the calculated value of motor corner by following formula

Calculated value with speed

$\widetilde{\mathrm{\θ}}\left(k\right)=\frac{2\mathrm{\πP}}{3}N\left(k\right)$

$\widetilde{\mathrm{\ω}}\left(k\right)=\frac{2\mathrm{\πP}}{3}(N\left(k\right)-N(k-1))/T$

In formula, k is the circulating sampling number of times, The k time calculated value of motor corner,

Be the k time calculated value of motor speed, N (k) is the total number of P-pulse that the k time detection obtains, and P is the magnetic pole logarithm of motor, and T is sense cycle;

(2) by following formula, carry out low-pass filtering treatment, obtain corner value θ and the velocity amplitude ω of motor;

$\mathrm{\θ}\left(k\right)=\mathrm{\α}\widetilde{\mathrm{\θ}}\left(k\right)+(1-\mathrm{\α})\mathrm{\θ}(k-1)$

$\mathrm{\ω}\left(k\right)=\mathrm{\α}\widetilde{\mathrm{\ω}}\left(k\right)+(1-\mathrm{\α})\mathrm{\ω}(k-1)$

In formula, θ (k) is the k time filtering output value of motor corner, and ω (k) is the k time filtering output value of motor speed, and α is filter coefficient, between 0 and 1;

(3) according to the dynamics of motor, build motor corner and speed observation model as follows:

$\widehat{\mathrm{\θ}}\left(k\right)=\widehat{\mathrm{\θ}}(k-1)+\widehat{\mathrm{\ω}}(k-1)*T$

$\widehat{\mathrm{\ω}}\left(k\right)=\frac{1}{J}(K_m{i}_q\left(k\right)-T_{\mathrm{fz}}\left(k\right))*T+\widehat{\mathrm{\ω}}(k-1)$

In formula,

Instantaneous corner measured value, For the instantaneous velocity measured value, J is the moment of inertia of motor, K _mMoment coefficient, i _qBe the quadrature axis component of motor stator electric current, adopt the command value of stator current quadrature axis component

Replace actual i _q, T _FzIt is load torque; By the load observation model, calculate, described load observation model is as follows:

$T_{\mathrm{fz}}\left(k\right)=a_1(\mathrm{\θ}\left(k\right)-\widehat{\mathrm{\θ}}\left(k\right))+a_2(\mathrm{\ω}\left(k\right)-\widehat{\mathrm{\ω}}\left(k\right))$

In formula, a ₁And a ₂Be the Load Torque Observer coefficient, but calculated off-line obtain;

Step 2, motor corner and speed observation model and the integration of load observation model by step 1 is set up, obtain corner, speed observer, for obtaining in real time the corner measured value

With the speed measured value

Be applied to control method as follows:

With the corner measured value For position feed back signal, by the given signal θ in position ^*With position feed back signal Subtract each other and obtain location following error signal, the given signal ω of location following error signal output speed after positioner ^*

With the speed measured value

For feedback speed signal, by speed preset signal ω ^*With feedback speed signal

Subtract each other and obtain speed error signal, speed error signal is exported the given signal of q shaft current after speed control

Detect the motor stator current i _a, i _b, form the quadrature axis current i through rotary converter _qWith direct-axis current i _dUsing the corner measured value as the rotation transformation angle of rotary converter, by the given signal of q shaft current

With the quadrature axis current i _qSubtract each other and obtain q shaft current error signal, this error signal is exported quadrature-axis voltage u after q shaft current controller _q, the given signal of d shaft current is 0, by direct-axis current i _dOutput direct-axis voltage u after d shaft current controller _d,

Described quadrature-axis voltage u _qWith direct-axis voltage u _dThrough contrary rotary converter, form motor stator three-phase alternating current pressure reference signal u _a, u _b, u _c, then through the SPWM modulation, for controlling the inverter of major loop, drive motors.

The factor alpha of the filter in described step 2 should be adjusted along with the variation of speed, and motor speed is lower, and the α value should be less, and speed is higher, and the α value should be larger; The factor alpha of described filter is less than or equal to 0.9 and be more than or equal to 0.1.

Technique effect

The present invention has provided a kind of sine wave vector control method of brushless DC motor, it only utilizes easy Hall switch position detector, the observer of realizing by software, reconstruct more accurate motor corner and the measured value of motor speed, utilize measured value to realize the sine wave vector control of brushless DC motor.This method had both retained the low advantage of the simple and reliable cost of brushless motor control system, had improved simultaneously control precision and dynamic property.

The accompanying drawing explanation

Fig. 1 is the structured flowchart of corner, speed observer;

Fig. 2 is based on the structured flowchart of the brushless DC motor control system of corner, speed observer;

Fig. 3 is the circuit theory diagrams of brushless DC motor control system;

Fig. 4 is brushless DC motor control system software main program flow chart;

Fig. 5 is brushless DC motor control system software interrupt service routine flow chart.

Specific embodiment

A kind of brshless DC motor corner and speed observation model and load observation model construction method are as follows:

(1) the sense cycle T to fix, read the three road position pulse signals that the motor hall detector is sent in real time, in real time the total number of Ji Qu tri-road P-pulse.By following formula, obtain the calculated value of motor corner

Calculated value with speed

$\widetilde{\mathrm{\θ}}\left(k\right)=\frac{2\mathrm{\πP}}{3}N\left(k\right)$

$\widetilde{\mathrm{\ω}}\left(k\right)=\frac{2\mathrm{\πP}}{3}(N\left(k\right)-N(k-1))/T$

In formula, k is the circulating sampling number of times,

The k time calculated value of motor corner,

Be the k time calculated value of motor speed, N (k) is the P-pulse number that the k time detection obtains, and P is the magnetic pole logarithm of motor, and T is sense cycle.

(2) the hall detector precision due to motor is low, and the stairstep signal that motor corner calculated value and speed calculated value are all, be discontinuous, must carry out low-pass filtering treatment by following formula, obtains corner value θ and the velocity amplitude ω of motor.

$\mathrm{\θ}\left(k\right)=\mathrm{\α}\widetilde{\mathrm{\θ}}\left(k\right)+(1-\mathrm{\α})\mathrm{\θ}(k-1)$

$\mathrm{\ω}\left(k\right)=\mathrm{\α}\widetilde{\mathrm{\ω}}\left(k\right)+(1-\mathrm{\α})\mathrm{\ω}(k-1)$

In formula, θ (k) is the k time filtering output value of motor corner, and ω (k) is the k time filtering output value of motor speed.α is filter coefficient, in theory between 0 and 1.

(3) factor alpha of filter should be adjusted along with the variation of speed, and motor speed is lower, and the α value should be less, and speed is higher, and the α value should be larger, and practical experience shows, α is less than or equal to 0.9 and be more than or equal to 0.1.

(4) by above-mentioned steps, obtain corner value θ and the velocity amplitude ω of motor, still have larger fluctuation, and the hysteresis on life period, in order to obtain instantaneous corner measured value

With the instantaneous velocity measured value

Should, according to the dynamics of motor, build " motor corner and speed observation model " as follows:

$\widehat{\mathrm{\θ}}\left(k\right)=\widehat{\mathrm{\θ}}(k-1)+\widehat{\mathrm{\ω}}(k-1)*T$

$\widehat{\mathrm{\ω}}\left(k\right)=\frac{1}{J}(K_m{i}_q\left(k\right)-T_{\mathrm{fz}}\left(k\right))*T+\widehat{\mathrm{\ω}}(k-1)$

In formula, J is the moment of inertia of motor, K _mMoment coefficient, i _qThe quadrature axis component of motor stator electric current, T _FzIt is load torque.

(5) in the observation model of reconstruct, for the convenience of actual operation, the command value of available stator current quadrature axis component Replace actual i _q.

(6) in the observation model of reconstruct, load torque T _FzCan according to the θ obtained previously with

Poor, ω with

Poor, be calculated as follows and obtain, this formula is called " load observation model ".

$T_{\mathrm{fz}}\left(k\right)=a_1(\mathrm{\θ}\left(k\right)-\widehat{\mathrm{\θ}}\left(k\right))+a_2(\mathrm{\ω}\left(k\right)-\widehat{\mathrm{\ω}}\left(k\right))$

In formula, a ₁And a ₂Be the Load Torque Observer coefficient, but calculated off-line obtain.

(7) " motor corner and speed observation model " and " load observation model " integrated, can be obtained " corner, speed observer ", for obtaining in real time the corner measured value

With the speed measured value With

The vector calculus and the FEEDBACK CONTROL that can be used for motor.The structure of " corner, speed observer " is shown in accompanying drawing 1.

Hall signal obtains the calculated value of motor corner through counting, sampling and precomputation

Calculated value with speed

The calculated value of motor corner

With the speed calculated value

Through low pass filter, process, obtain corner value θ and the velocity amplitude ω of motor, dynamics according to motor, build " motor corner and speed observation model ", build simultaneously " load observation model ", " motor corner and speed observation model " and " load observation model " integrated, can be obtained " corner, speed observer ", Real-time Obtaining corner measured value

With the speed measured value

(8) by " corner, speed observer " for the brshless DC motor closed-loop control system, can realize that sine wave vector controls.Based on the structure of the brushless DC motor control system of " corner, speed observer " as shown in Figure 2.

With the corner measured value For position feed back signal, the given signal θ in position ^*With position feed back signal

Subtract each other and obtain the location following error, the location following error is through positioner, the given signal ω of output speed ^*, with the speed measured value

For feedback speed signal, speed preset signal ω ^*With feedback speed signal

Subtract each other and obtain velocity error, velocity error, through speed control, is exported the given signal of q shaft current Detect the motor stator current i _a, i _b, form the quadrature axis current i through rotary converter _qWith direct-axis current i _dThe rotation transformation angle of rotary converter is The given signal of q shaft current

With the quadrature axis current i _qSubtract each other and obtain q shaft current error, this error, through q shaft current controller, is exported quadrature-axis voltage u _q, the given signal of d shaft current is 0, direct-axis current i _dThrough d shaft current controller, output direct-axis voltage u _d, quadrature-axis voltage u _qWith direct-axis voltage u _d, form motor stator three-phase alternating current pressure reference signal u through contrary rotary converter _a, u _b, u _c, then through the SPWM modulation, for controlling the inverter of major loop, drive motors.

In the system shown in accompanying drawing 2, outer shroud is position ring, and the given signal in position is θ ^*, position feed back signal is

The output signal of positioner is speed preset signal ω ^*, feedback speed signal is

Speed control adopts the PI control algolithm.

The inside of speed ring, comprised q shaft current ring and d shaft current ring, and the output signal of speed control is the given signal of q shaft current

The given signal of d shaft current is 0.Detect the motor stator current i _a, i _b, form the quadrature axis current i through rotation transformation _qWith direct-axis current i _d, for feedback.The rotation transformation angle is

From " corner, speed observer ".

Q shaft current controller and d shaft current controller all adopt the PI control algolithm, and its output is respectively quadrature-axis voltage u _qWith direct-axis voltage u _d, can form motor stator three-phase alternating current pressure reference signal u through contrary rotation transformation _a, u _b, u _c, then through the SPWM modulation, for controlling the inverter of major loop, drive motors.

(9) as shown in Figure 3, the processor of native system is to adopt TMS320F28035, the modules such as the corner in accompanying drawing 2, speed observer, positioner, speed control, d shaft current controller, q shaft current controller, rotation transformation, contrary rotation transformation realize by software, and the SVPWM module is realized by the hardware of TMS320F28035 inside.Adopt two current sensor ACS712 to detect the motor stator current i _a, i _b, the signal of current sensor is input to the A/D translation interface of TMS320F28035.

Major loop is for handing over-straight-knot structure, and inverter is wherein realized by Intelligent Power Module IRAMX20UP60A.

As shown in Figure 4 and Figure 5, be respectively brshless DC motor and control software main flow chart and brushless DC motor control system software interrupt service routine flow chart.It is as follows that described brshless DC motor is controlled the software main flow: 1) initialization system clock, interrupt vector table, IO port, 2) parameter list is controlled in initialization, 3) initialization A/D change-over circuit, PWM output, 4) start the control cycle timer, 5) wait for that the control cycle timer interrupts, refreshes man-machine interface, 6) read Hall element signal Ha, Hb, Hc, calculate the motor corner

And rotating speed

, 7) and calculating motor corner filter value θ (k) and rotating speed filter value ω (k), 8) calculating corner measured value

With the speed measured value

, 9) and according to current i _a, i _b, i _cCurrency, calculate i _q(k) and i _d(k); Respectively with 1 times, 2 times, the 4 times control cycles in major cycle cycle, completing place controller, speed control, current controller refresh computing; 10) according to handing over, the output signal u of direct-axis current controller _q(k) and u _dAnd the corner measured value (k),

, by contrary rotation transformation, calculate the reference value of Three-phase electronic voltage, refresh the SVPWM link.Brushless DC motor control system software interrupt service routine is as follows: 1) response Interruption, 2) start A/D converter, 3) read A phase current sampling value, 4) read B phase current sampling value, 5) read C phase current sampling value, 6) interrupt returning.

Claims (2)

1. control method for brushless direct current motor is characterized in that: comprise the following steps:

Step 1, set up corner, speed and the load observation model of brshless DC motor, specific as follows:

(1) the sense cycle T to fix, read the three road position pulse signals that the motor hall detector is sent in real time, and the real-time total number of Ji Qu tri-road P-pulse obtains the calculated value of motor corner by following formula

Calculated value with speed

$\widetilde{\mathrm{\θ}}\left(k\right)=\frac{2\mathrm{\πP}}{3}N\left(k\right)$

$\widetilde{\mathrm{\ω}}\left(k\right)=\frac{2\mathrm{\πP}}{3}(N\left(k\right)-N(k-1))/T$

In formula, k is the circulating sampling number of times,

The k time calculated value of motor corner,

Be the k time calculated value of motor speed, N (k) is the total number of P-pulse that the k time detection obtains, and P is the magnetic pole logarithm of motor, and T is sense cycle;

(2) by following formula, carry out low-pass filtering treatment, obtain corner value θ and the velocity amplitude ω of motor;

$\mathrm{\θ}\left(k\right)=\mathrm{\α}\widetilde{\mathrm{\θ}}\left(k\right)+(1-\mathrm{\α})\mathrm{\θ}(k-1)$

$\mathrm{\ω}\left(k\right)=\mathrm{\α}\widetilde{\mathrm{\ω}}\left(k\right)+(1-\mathrm{\α})\mathrm{\ω}(k-1)$

In formula, θ (k) is the k time filtering output value of motor corner, and ω (k) is the k time filtering output value of motor speed, and α is filter coefficient, between 0 and 1;

(3) according to the dynamics of motor, build motor corner and speed observation model as follows:

$\widehat{\mathrm{\θ}}\left(k\right)=\widehat{\mathrm{\θ}}(k-1)+\widehat{\mathrm{\ω}}(k-1)*T$

$\widehat{\mathrm{\ω}}\left(k\right)=\frac{1}{J}(K_m{i}_q\left(k\right)-T_{\mathrm{fz}}\left(k\right))*T+\widehat{\mathrm{\ω}}(k-1)$

In formula,

Instantaneous corner measured value,

For the instantaneous velocity measured value, J is the moment of inertia of motor, K _mMoment coefficient, i _qBe the quadrature axis component of motor stator electric current, adopt the command value of stator current quadrature axis component

Replace actual i _q, T _FzIt is load torque; By the load observation model, calculate, described load observation model is as follows:

$T_{\mathrm{fz}}\left(k\right)=a_1(\mathrm{\θ}\left(k\right)-\widehat{\mathrm{\θ}}\left(k\right))+a_2(\mathrm{\ω}\left(k\right)-\widehat{\mathrm{\ω}}\left(k\right))$

In formula, a ₁And a ₂Be the Load Torque Observer coefficient, but calculated off-line obtain;

Step 2, motor corner and speed observation model and the integration of load observation model by step 1 is set up, obtain corner, speed observer, for obtaining in real time the corner measured value

With the speed measured value

Be applied to control method as follows:

With the corner measured value

For position feed back signal, by the given signal θ in position ^*With position feed back signal Subtract each other and obtain location following error signal, the given signal ω of location following error signal output speed after positioner ^*

With the speed measured value

For feedback speed signal, by speed preset signal ω ^*With feedback speed signal

Subtract each other and obtain speed error signal, speed error signal is exported the given signal of q shaft current after speed control

Detect the motor stator current i _a, i _b, form the quadrature axis current i through rotary converter _qWith direct-axis current i _dUsing the corner measured value as the rotation transformation angle of rotary converter, by the given signal of q shaft current

With the quadrature axis current i _qSubtract each other and obtain q shaft current error signal, this error signal is exported quadrature-axis voltage u after q shaft current controller _q, the given signal of d shaft current is 0, by direct-axis current i _dOutput direct-axis voltage u after d shaft current controller _d,

Described quadrature-axis voltage u _qWith direct-axis voltage u _dThrough contrary rotary converter, form motor stator three-phase alternating current pressure reference signal u _a, u _b, u _c, then through the SPWM modulation, for controlling the inverter of major loop, drive motors.

2. control method for brushless direct current motor according to claim 1, it is characterized in that: the factor alpha of the filter in described step 2 should be adjusted along with the variation of speed, and motor speed is lower, and the α value should be less, and speed is higher, and the α value should be larger; The factor alpha of described filter is less than or equal to 0.9 and be more than or equal to 0.1.

Images