CN101242154A - A built-in permanent magnetic brushless DC motor control system for no position sensor - Google Patents

Abstract

The invention relates to an embedded permanent brushless DC motor non-position sensor control system, composed of a microprocessor controller, a IPM module and a drive unit, a current detection circuit, a phase voltage detection circuit, an output display circuit, a control input circuit, an embedded rotor structured permanent brushless DC motor, a drive and control power source, etc. The invention judges the position of rotor by detecting variation of inductance of stator winding indirectly according to the character that the inductance of rotor winding of the embedded rotor structured permanent brushless DC motor is the function of rotor position theta. In the embedded rotor structured permanent brushless DC motor with star-shaped connection adopted by stator winding, when self inductance of two phase windings therein are equal, the position is the counter potential zero-crossing point without phase connection. Therefore, adopting non-connection phase voltage to ground continuously in one PWM period, and comparing the difference value, to judge non-connection phase counter potential zero-crossing point, and lagging behind for 30 degrees electric angel as phase-exchange time of next time. The method of the invention is indirect induction method. The non-position sensor control method of indirection induction method is nothing to do with rotating speed of motor, so that the invention overcomes the problem of malfunction at low speed of the conventional non-position sensor control method-the detection method of counter potential zero-crossing point. The invention can work reliably at ultra-low speed even if the rotating speed of motor is close to zero.

Description

A kind of built-in permanent magnetic brushless DC motor control system of position-sensor-free

Technical field

The present invention relates to a kind of brushless DC motor control system of position-sensor-free, particularly have the control system without position sensor of permanent-magnet brushless DC electric machine of the star-like connection of embedded rotor structure.

Background technology

The operation of permanent-magnet brushless DC electric machine is correspondingly to change its different assembled state that trigger by power device of inverter with the diverse location of rotor to realize.Therefore accurately the position of detection rotor, and be the key of the normal operation of control brshless DC motor according to the triggering assembled state of the punctual power switched device of rotor-position.The use location transducer is the most direct effective method as the position detecting device of rotor.Position transducer is one of part of brushless direct current motor system, its effect is the position of detection rotor in motion process, the position signalling of rotor magnetic steel magnetic pole is changed into the signal of telecommunication, for logic switching circuit provides correct commutation information, with the conducting of controlling them with end, make electric current in the armature winding along with the variation commutation in order of rotor-position, form step-type rotating magnetic field in the air gap, drive p-m rotor and continuously rotate.Generally position transducer is installed on the axle of rotor, realizes the real-time detection of rotor-position.Position transducer the earliest is a magneto-electric, and not only heaviness but also complexity are eliminated, and the hall position sensor of Mageneto-sensitive type is widely used in the brushless DC motor at present, also has the position transducer of photoelectric type in addition.The existence of position transducer has increased the volume of motor, and transducer has wearing and tearing unavoidably during rotation, and needs many holding wires, has more reduced the reliability of system, has also increased technological requirement and cost that motor is made.In addition, because of can't installation position and velocity transducer, and because some shortcomings of transducer self, and factor such as system cost, impel people that the position Sensorless Control of brushless DC motor is launched research in some occasion.The position Sensorless Control technology has been simplified the motor body structure owing to saved position transducer, has obtained good effect.The position Sensorless Control Study on Technology has great importance at some special occasions such as high-speed electric expreess locomotive, micromachine, Aero-Space, underwater robot, household electrical appliance.

If in X in brushless DC motor control without position transducer, just must be by means of to the detection of relevant amount with calculate to obtain the position of motor rotor with motor rotor position.At present brushless DC motor does not have transducer and controls comparatively that the typical control method has back electromotive force method, fly-wheel diode method, stator triple-frequency harmonics method and adopts Based Intelligent Control detection method etc.Wherein the back electromotive force method is the most common and most widely used method, what for example United States Patent (USP) NO.5235264 adopted is exactly the back electromotive force method, its main feature is directly to detect the back electromotive force zero-crossing of responding in the no power phase stator winding in the motor, and the 30 ° of electrical degrees that lag behind again are the commutation point.Because the back electromotive force of responding in the stator winding is directly proportional with speed, back electromotive force is very faint when motor speed is very low, just can't judge the zero crossing of back electromotive force, so that this method tends to when motor low speed is invalid.And Chinese patent ZL200410065332.1 discloses " a kind of brshless DC motor first-harmonic method direct Torque Control based on the position-sensor-free technology ", and it calculates rotating speed with the transient current, the space voltage vector of selecting for use that detect and the magnetic linkage that calculates.But this method must be set up the method for rotating speed Mathematical Modeling accurately, because the complexity of alternating current machine itself, setting up accurately, the rotating speed Mathematical Modeling is very difficult thing.The present invention is different with above method, do not need to set up the Mathematical Modeling of complicated motor, utilize the method for measuring the inductance of motor stator indirectly to determine the zero crossing of back electromotive force, it only with the relating to parameters of motor self, irrelevant with rotating speed of motor, can when motor speed is extremely low, work reliably.

Summary of the invention

The objective of the invention is: in order to solve the problem that method for controlling position-less sensor commonly used-" back electromotive force method " lost efficacy in the low speed territory or near the zero-speed territory, and provide a kind of working range that can enlarge position Sensorless Control, the built-in permanent magnetic brushless DC motor control system of position-sensor-free that also can reliably working when utmost point low speed (rated speed 5%).

The technical solution that the present invention taked: a kind of built-in permanent magnetic brushless DC motor control system of position-sensor-free, hardware comprises the permanent-magnet brushless DC electric machine of microprocessor controller, IPM module and driver element and embedded rotor structure as shown in Figure 9; It is characterized in that: microprocessor controller calculates the position signalling of rotor according to detected voltage, current signal, produces to comprise six road pwm control signals; After pwm control signal process IPM module and the driver element conversion, produce the high-voltage signal that width changes with PWM, drive IPM module circulation conducting, make the permanent-magnet brushless DC electric machine running of embedded rotor structure.The given rotating speed of input can come selection speed from the given of host computer or by a potentiometer by serial communication; The bus current of its input is via microcontroller control circuit sampling processing, to protect and motor operating parameter as over current of motor.Microprocessor controller adopts the method for " continuous quadratic " location to give the rotor pre-determined bit during control system operation, has determined the initial position of rotor; T1 in same PWM period T and the t2 no power voltage to earth mutually of sampling constantly, and calculate this difference of twice is if the inferior difference of N continuous (3～5) is then assert the back electromotive force zero-crossing of this phase less than set point; According to the 60 ° of electrical degrees of being separated by of twice adjacent back electromotive force zero-crossing in the last circulation, thereby calculate 30 ° of electrical degrees, the commutation that the 30 ° of electrical degrees that lag behind again on the basis of back-emf zero crossing are next phase constantly; Motor has whenever circled 6 commutations constantly, just can calculate the mean angular velocity during the commutation interval 2 times as long as record 2 commutations time interval constantly; The set-point of electric current loop carries out the PI computing by the rotating speed deviation en to der Geschwindigkeitkreis and obtains, and the deviate ei of electric current reconciles computing by current PI and becomes PWM ripple duty ratio, thereby controls the rotating speed of the permanent-magnet brushless DC electric machine of embedded rotor structure.

Principle of the present invention is: the motor body in the permanent-magnet brushless DC electric machine is a permanent magnet synchronous motors, and when rotor was embedded structure, motor had the characteristics of salient pole machine.The inductance of salient pole synchronous electric machine stator winding is the function of motor rotor position, according to these characteristics, can judge the position of rotor by the inductance size of detected stator winding, thereby judge the zero crossing of the back electromotive force of no power phase, the 30 ° of electrical degrees that lag behind again are the commutation point.

Can obtain the expression formula of motor three phase winding self-inductions and mutual inductance according to the Mathematical Modeling of salient pole machine:

L _aa＝L _aa0+L _al+L _g2 cos(2θ) (1)

$L_{\mathrm{bb}}=L_{\mathrm{aa}0}+L_{\mathrm{al}}+{\mathrm{<figure-callout\; id="2"\; label="L\; g"\; filenames="S2008100694727E00032.png"\; state="\{\{state\}\}">L</figure-callout>}}_g\cos (2\mathrm{\&theta;}+\frac{2\mathrm{\&pi;}}{3})---\left(2\right)$

$L_{\mathrm{cc}}=L_{\mathrm{aa}0}+L_{\mathrm{al}}+{\mathrm{<figure-callout\; id="2"\; label="L\; g"\; filenames="S2008100694727E00032.png"\; state="\{\{state\}\}">L</figure-callout>}}_g\cos (2\mathrm{\&theta;}-\frac{2\mathrm{\&pi;}}{3})---\left(3\right)$

$L_{\mathrm{ab}}=L_{\mathrm{ba}}={-0.5L}_{\mathrm{aa}0}+{\mathrm{<figure-callout\; id="2"\; label="L\; g"\; filenames="S2008100694727E00032.png"\; state="\{\{state\}\}">L</figure-callout>}}_g\cos (2\mathrm{\&theta;}-\frac{2\mathrm{\&pi;}}{3})---\left(4\right)$

L _bc＝L _cb＝-0.5L _aa0+L _g2 cos(2θ) (5)

$L_{\mathrm{ac}}=L_{\mathrm{ca}}={-0.5L}_{\mathrm{aa}0}+{\mathrm{<figure-callout\; id="2"\; label="L\; g"\; filenames="S2008100694727E00032.png"\; state="\{\{state\}\}">L</figure-callout>}}_g\cos (2\mathrm{\&theta;}+\frac{2\mathrm{\&pi;}}{3})---\left(6\right)$

L wherein _Aa, L _Bb, L _CcBe A, B, the self-induction of C three-phase; L _Ac, L _Bc, L _AcBe A, B, the C three-phase for mutual inductance.

Can draw table 1 according to (1)～(6) formula and Fig. 2:

Table 1-stator winding self-induction and the situation of mutual inductance in the time of 0 °～330 °

Self-induction	Lcc ＝ Lbb	Lcc ＝ Laa	Laa ＝ Lbb	Lcc ＝ Lbb	Laa ＝ Lcc	Lbb ＝ Laa	Lcc ＝ Lbb	Lcc ＝ Laa	Laa ＝ Lbb	Lcc ＝ Lbb	Laa ＝ Lcc	Lbb ＝ Laa
													Mutual inductance	Lab ＝ Lac	Lab ＝ Lbc	Lab ＝ Lac	Lab ＝ Lac	Lab ＝ Lbc	Lab ＝ Lac	Lab ＝ Lac	Lab ＝ Lbc	Lab ＝ Lac	Lab ＝ Lac	Lab ＝ Lbc	Lab ＝ Lac
Angle	0	30	60	90-	120	150	180	210	240	270	300	330

As shown in Figure 1, the permanent-magnet brushless DC electric machine of star-like connection works in " two or two conducting " mode, and 6 operating states were arranged in each cycle, is equivalent to the working method of a direct current machine in each state.Use for reference the working method of direct current machine H bridge bipolarity copped wave, allow permanent-magnet brushless DC electric machine work in mode of operation-bipolarity chopping way of Fig. 3.When duty ratio was 50%, the bus output voltage was 0 as shown in Figure 3.Represent phase under the conduction device to flow to load end with capitalization below, represent to flow to power supply, promptly use a+ with subscript "-" with subscript "+" expression, a-, b+, b-, c+ and c-represent the conducting situation of 6 IGBT pipes respectively.Fig. 4 is the a+b-conductive path, and Fig. 5 is the a-b+ conductive path.By Fig. 6, Fig. 7 as can be known under two kinds of conduction modes no power winding c voltage to earth be:

A+b-conduction mode: Vcy=Vcs+Vsy

A-b+ conduction mode: V ' cy=Vcs '+Vs ' y

At the a+b-conduction mode:

$v_a=L_{\mathrm{aa}}\frac{{\mathrm{di}}_a}{\mathrm{dt}}+L_{\mathrm{ab}}\frac{{\mathrm{di}}_b}{\mathrm{dt}}+i_a\frac{{\mathrm{dL}}_{\mathrm{ab}}}{\mathrm{dt}}+i_b\frac{{\mathrm{dL}}_{\mathrm{ab}}}{\mathrm{dt}}+i_{a}R+E_a---\left(7\right)$

$v_b=L_{\mathrm{ab}}\frac{{\mathrm{di}}_a}{\mathrm{dt}}+L_{\mathrm{bb}}\frac{{\mathrm{di}}_b}{\mathrm{dt}}+i_a\frac{{\mathrm{dL}}_{\mathrm{ab}}}{\mathrm{dt}}+i_b\frac{{\mathrm{dL}}_{\mathrm{bb}}}{\mathrm{dt}}+i_{b}R+E_b---\left(8\right)$

Because: i _a=-i _b

$v_b=(L_{\mathrm{ab}}-L_{\mathrm{bb}})\frac{{\mathrm{di}}_a}{\mathrm{dt}}+i_a\frac{d(L_{\mathrm{ab}}-L_{\mathrm{bb}})}{\mathrm{dt}}+i_{a}R+E_b---\left(9\right)$

If consider the IGBT tube voltage drop of two conductings, can get: V wherein _IGBTTube voltage drop during for the IGBT conducting

V _dc＝v _a-v _b+2V _IGBT

v _sy＝-v _b+V _IGBT

The above equation of simultaneous can get:

$V_{\mathrm{dc}}=(L_{\mathrm{aa}}+L_{\mathrm{bb}}-{2L}_{\mathrm{ab}})\frac{{\mathrm{di}}_a}{\mathrm{dt}}+i_a\frac{d(L_{\mathrm{aa}}+L_{\mathrm{bb}}-{2L}_{\mathrm{ab}})}{\mathrm{dt}}+{2i}_{a}R+(E_a-E_b)+{2V}_{\mathrm{IGBT}}---\left(10\right)$

According to table 1 and Fig. 2 as can be known, when AB or BA conducting, the θ angle equals: 60,240 o'clock

L is arranged _Aa=L _Bb, E _a=-E _b(11)

The neutral point voltage to earth is:

$v_{\mathrm{sy}}=\frac{V_{\mathrm{dc}}}{2}---\left(12\right)$

v _cs＝L _aci _a+L _bci _b+E _c＝i _a(L _ac-L _bc)+E _c (13)

At this moment C phase-to-ground voltage: $v_{\mathrm{cy}}=v_{\mathrm{cs}}+v_{\mathrm{sy}}=\frac{{\mathrm{<figure-callout\; id="2"\; label="V\; dc"\; filenames="S2008100694727E00032.png"\; state="\{\{state\}\}">V</figure-callout>}}_{\mathrm{dc}}}{2}+i_a(L_{\mathrm{ac}}-L_{\mathrm{bc}})+E_c---\left(14\right)$

Same derivation, under the a-b+ conduction mode:

${v_a}^{\&prime;}=L_{\mathrm{aa}}\frac{{{\mathrm{di}}_a}^{\&prime;}}{\mathrm{dt}}+L_{\mathrm{ab}}\frac{{{\mathrm{di}}_b}^{\&prime;}}{\mathrm{dt}}{+i_a}^{\&prime;}\frac{{\mathrm{dL}}_{\mathrm{ab}}}{\mathrm{dt}}{+i_b}^{\&prime;}\frac{{\mathrm{dL}}_{\mathrm{ab}}}{\mathrm{dt}}+{i_a}^{\&prime;}R+E_a---\left(15\right)$

${v_b}^{\&prime;}=L_{\mathrm{ab}}\frac{{{\mathrm{di}}_a}^{\&prime;}}{\mathrm{dt}}+L_{\mathrm{bb}}\frac{{{\mathrm{di}}_b}^{\&prime;}}{\mathrm{dt}}+{i_a}^{\&prime;}\frac{{\mathrm{dL}}_{\mathrm{ab}}}{\mathrm{dt}}{+i_b}^{\&prime;}\frac{{\mathrm{dL}}_{\mathrm{bb}}}{\mathrm{dt}}{+i_b}^{\&prime;}R+E_b---\left(16\right)$

i _a′＝-i _b′ (17)

If consider the pressure drop of the fly-wheel diode of two conductings, can get: V wherein _DTube voltage drop during for the fly-wheel diode conducting

V _dc＝v _b′-v _a′-2V _D (18)

$V_{\mathrm{dc}}=-2(L_{\mathrm{aa}}-L_{\mathrm{bb}})\frac{{{\mathrm{di}}_a}^{\&prime;}}{\mathrm{dt}}-{{2i}_a}^{\&prime;}\frac{d(L_{\mathrm{aa}}-L_{\mathrm{ab}})}{\mathrm{dt}}-{{2i}_a}^{\&prime;}R-{2E}_b-{2V}_D---\left(19\right)$

v _s′y＝-v _a′-V _D (20)

By table 1 and Fig. 2, when AB or BA conducting, the θ angle equals:, L is arranged at 60,240 o'clock _Aa=L _Bb, E _a=-E _b:

$v_{s^{\&prime;}y}=\frac{V_{\mathrm{dc}}}{2}---\left(21\right)$

v _cs′＝L _aci _a′+L _bci _b′+E _c＝i _a′(L _ac-L _bc)+E _c (22)

At this moment C phase-to-ground voltage: v _Cy'=v _Cs'+v _Sy' (23)

The θ angle equals: 60,240 o'clock mutual inductance: L _Ac=L _Bc, v _Cs=v _Cs'

Existing: the θ angle equals: 60,240 o'clock: v _Cy=v _Cy', if not when these angles, v then _Cy≠ v _Cy'

Below further analyze the θ angle and be not equal to: 60,240 o'clock, the voltage to earth situation of change of non-conductive phase C under the bipolarity copped wave pattern:

Under the a+b-conduction mode:

$\frac{{\mathrm{<figure-callout\; id="2"\; label="V\; dc"\; filenames="S2008100694727E00032.png"\; state="\{\{state\}\}">V</figure-callout>}}_{\mathrm{dc}}}{2}=(\frac{L_{\mathrm{aa}}+L_{\mathrm{bb}}}{2}-L_{\mathrm{ab}})\frac{{\mathrm{di}}_a}{\mathrm{dt}}+i_a\frac{d(\frac{L_{\mathrm{aa}}+L_{\mathrm{bb}}}{2}-L_{\mathrm{ab}})}{\mathrm{dt}}+i_{a}R+\frac{E_a-{\mathrm{<figure-callout\; id="2"\; label="E\; b"\; filenames="S2008100694727E00032.png"\; state="\{\{state\}\}">E</figure-callout>}}_b}{2}+V_{\mathrm{IGBT}}---\left(24\right)$

$v_{\mathrm{sy}}=\frac{V_{\mathrm{<figure-callout\; id="2"\; label="dc"\; filenames="S2008100694727E00032.png"\; state="\{\{state\}\}">dc</figure-callout>}}}{2}+\left(\frac{L_{\mathrm{bb}}-L_{\mathrm{aa}}}{2}\right)\frac{{\mathrm{di}}_a}{\mathrm{dt}}+i_a\frac{d\left(\frac{L_{\mathrm{bb}}-L_{\mathrm{aa}}}{2}\right)}{\mathrm{dt}}-\frac{(E_a+E_b)}{2}---\left(25\right)$

And under the a-b+ conduction mode:

$\frac{{\mathrm{<figure-callout\; id="2"\; label="V\; dc"\; filenames="S2008100694727E00032.png"\; state="\{\{state\}\}">V</figure-callout>}}_{\mathrm{dc}}}{2}=-(\frac{L_{\mathrm{aa}}+L_{\mathrm{bb}}}{2}-L_{\mathrm{ab}})\frac{{{\mathrm{di}}_a}^{\&prime;}}{\mathrm{dt}}-{i_a}^{\&prime;}\frac{d(\frac{L_{\mathrm{aa}}+L_{\mathrm{bb}}}{2}-L_{\mathrm{ab}})}{\mathrm{dt}}-{i_a}^{\&prime;}R-\frac{(E_a-E_b)}{2}-V_D---\left(26\right)$

$v_{s\&prime;y}=\frac{V_{\mathrm{dc}}}{2}-\left(\frac{L_{\mathrm{aa}}-L_{\mathrm{bb}}}{2}\right)\frac{{{\mathrm{di}}_a}^{\&prime;}}{\mathrm{dt}}-{i_a}^{\&prime;}\frac{d\left(\frac{L_{\mathrm{aa}}-L_{\mathrm{bb}}}{2}\right)}{\mathrm{dt}}-\frac{(E_a+E_b)}{2}---\left(27\right)$

Compare (25) and (27) as can be known: central point S point voltage to earth exists

Near fluctuation.

Below further analyze the situation of change of fluctuation:

When the a+b-conduction mode: according to formula (24)

$\frac{{\mathrm{di}}_a}{\mathrm{dt}}=\frac{(\frac{V_{\mathrm{dc}}}{2}-i_a\frac{d(\frac{L_{\mathrm{aa}}+L_{\mathrm{bb}}}{2}-L_{\mathrm{ab}})}{\mathrm{dt}}-\frac{E_a-E_b}{2}-V_{\mathrm{IGBT}}-i_{a}R)}{(\frac{L_{\mathrm{aa}}+L_{\mathrm{bb}}}{2}-L_{\mathrm{ab}})}---\left(28\right)$

When the a-b+ state: have according to formula (26):

$\frac{{{\mathrm{di}}_a}^{\&prime;}}{\mathrm{dt}}=\frac{-(\frac{{\mathrm{<figure-callout\; id="2"\; label="V\; dc"\; filenames="S2008100694727E00032.png"\; state="\{\{state\}\}">V</figure-callout>}}_{\mathrm{dc}}}{2}+{i_a}^{\&prime;}\frac{d(\frac{L_{\mathrm{aa}}+L_{\mathrm{bb}}}{2}-L_{\mathrm{ab}})}{\mathrm{dt}}+\frac{E_a+{\mathrm{<figure-callout\; id="2"\; label="E\; b"\; filenames="S2008100694727E00032.png"\; state="\{\{state\}\}">E</figure-callout>}}_b}{2}+V_D+{i_a}^{\&CenterDot;}R)}{(\frac{L_{\mathrm{aa}}+L_{\mathrm{bb}}}{2}-L_{\mathrm{ab}})}---\left(29\right)$

Fig. 8 is carried out labor, analyzes its typical conditions in a PWM period T:

Work as duty ratio $\mathrm{\&rho;}=\frac{T0}{T}\&GreaterEqual;50\%,$ Order: Δ v _Sy(T)=v _Sy(t1)-v _{S ' y}(t2)

${\mathrm{\&Delta;v}}_{\mathrm{sy}}\left(T\right)=(\frac{{\mathrm{<figure-callout\; id="2"\; label="V\; dc"\; filenames="S2008100694727E00032.png"\; state="\{\{state\}\}">V</figure-callout>}}_{\mathrm{dc}}}{2}+\left(\frac{L_{\mathrm{bb}}-L_{\mathrm{aa}}}{2}\right)\frac{{\mathrm{di}}_a}{\mathrm{dt}}+i_a\frac{d\left(\frac{L_{\mathrm{bb}}-L_{\mathrm{aa}}}{2}\right)}{\mathrm{dt}}-\frac{(E_a+E_b)}{2}){|}_{t=\mathrm{<figure-callout\; id="1"\; label="t"\; filenames="S2008100694727E00031.png,S2008100694727E00032.png"\; state="\{\{state\}\}">t</figure-callout>}1}$

$-(\frac{V_{\mathrm{dc}}}{2}-\left(\frac{L_{\mathrm{aa}}-L_{\mathrm{bb}}}{2}\right)\frac{{{\mathrm{di}}_a}^{\&prime;}}{\mathrm{dt}}-{i_a}^{\&prime;}\frac{d\left(\frac{L_{\mathrm{aa}}-L_{\mathrm{bb}}}{2}\right)}{\mathrm{dt}}-\frac{(E_a+E_b)}{2}){|}_{t=\mathrm{<figure-callout\; id="2"\; label="t"\; filenames="S2008100694727E00032.png"\; state="\{\{state\}\}">t</figure-callout>}2}$

$=(\left(\frac{L_{\mathrm{bb}}-L_{\mathrm{aa}}}{2}\right)(\frac{{\mathrm{di}}_a}{\mathrm{dt}}{|}_{t=t1}-\frac{{{\mathrm{di}}_a}^{\&prime;}}{\mathrm{dt}}{|}_{t=t2})+\frac{d\left(\frac{L_{\mathrm{bb}}-L_{\mathrm{aa}}}{2}\right)}{\mathrm{dt}}(i_a{|}_{t=t1}-{i_a}^{\&prime;}{|}_{t=t2}))---\left(30\right)$

Respectively will

With Bring formula (30) into, in Fig. 8, notice i _a(t1)=i _a' (t2), under the sufficiently high situation of chopping frequency (frequency is greater than 10K):

${\mathrm{\&Delta;v}}_{\mathrm{sy}}\left(T\right)=\left(\frac{L_{\mathrm{bb}}-L_{\mathrm{aa}}}{2}\right)\left(\frac{V_{\mathrm{dc}}-V_{\mathrm{IGBT}}+V_D}{\frac{L_{\mathrm{aa}}+L_{\mathrm{bb}}}{2}-L_{\mathrm{ab}}}\right)---\left(31\right)$

Because: V _DcV _IGBT-V _D

$\left(\frac{L_{\mathrm{bb}}-L_{\mathrm{aa}}}{2}\right)=\frac{\sqrt{3}}{2}{\mathrm{<figure-callout\; id="2"\; label="L\; g"\; filenames="S2008100694727E00032.png"\; state="\{\{state\}\}">L</figure-callout>}}_g\cos (2\mathrm{\&theta;}+\frac{5\mathrm{\&pi;}}{6})---\left(32\right)$

So: ${\mathrm{\&Delta;v}}_{\mathrm{sy}}=\frac{V_{\mathrm{dc}}\left(\frac{\sqrt{3}}{2}{\mathrm{<figure-callout\; id="2"\; label="L\; g"\; filenames="S2008100694727E00032.png"\; state="\{\{state\}\}">L</figure-callout>}}_g\cos (2\mathrm{\&theta;}+\frac{5\mathrm{\&pi;}}{6})\right)}{\frac{L_{\mathrm{aa}}+L_{\mathrm{bb}}}{2}-L_{\mathrm{ab}}}---\left(33\right)$

$L_d=L_{\mathrm{al}}+\frac{3}{2}({\mathrm{<figure-callout\; id="0"\; label="L\; aa"\; filenames="S2008100694727E00012.png,S2008100694727E00051.png"\; state="\{\{state\}\}">L</figure-callout>}}_{\mathrm{aa}}+L_{g2})$

$L_q=L_{\mathrm{al}}+\frac{3}{2}(L_{\mathrm{aa}0}-L_{g2})$

$\frac{L_{\mathrm{aa}}+L_{\mathrm{bb}}}{2}-L_{\mathrm{ab}}=\frac{({\mathrm{<figure-callout\; id="0"\; label="L\; aa"\; filenames="S2008100694727E00012.png,S2008100694727E00051.png"\; state="\{\{state\}\}">L</figure-callout>}}_{\mathrm{aa}}+L_{\mathrm{al}}+{\mathrm{<figure-callout\; id="2"\; label="L\; g"\; filenames="S2008100694727E00032.png"\; state="\{\{state\}\}">L</figure-callout>}}_g\cos \left(2\mathrm{\&theta;}\right))+({\mathrm{<figure-callout\; id="0"\; label="L\; aa"\; filenames="S2008100694727E00012.png,S2008100694727E00051.png"\; state="\{\{state\}\}">L</figure-callout>}}_{\mathrm{aa}}+L_{\mathrm{al}}+{\mathrm{<figure-callout\; id="2"\; label="L\; g"\; filenames="S2008100694727E00032.png"\; state="\{\{state\}\}">L</figure-callout>}}_g\cos (2\mathrm{\&theta;}+\frac{2\mathrm{\&pi;}}{3}))}{2}-({-0.5\mathrm{<figure-callout\; id="0"\; label="L\; aa"\; filenames="S2008100694727E00012.png,S2008100694727E00051.png"\; state="\{\{state\}\}">L</figure-callout>}}_{\mathrm{aa}}+{\mathrm{<figure-callout\; id="2"\; label="L\; g"\; filenames="S2008100694727E00032.png"\; state="\{\{state\}\}">L</figure-callout>}}_g\cos (2\mathrm{\&theta;}-\frac{2\mathrm{\&pi;}}{3}))$

$=L_{\mathrm{al}}+{1.5\mathrm{<figure-callout\; id="0"\; label="L\; aa"\; filenames="S2008100694727E00012.png,S2008100694727E00051.png"\; state="\{\{state\}\}">L</figure-callout>}}_{\mathrm{aa}}+L_{g2}(\frac{\mathrm{<figure-callout\; id="2"\; label="cos"\; filenames="S2008100694727E00032.png"\; state="\{\{state\}\}">cos</figure-callout>}2\mathrm{\&theta;}+\cos (2\mathrm{\&theta;}+\frac{2\mathrm{\&pi;}}{3})}{2}-\cos (2\mathrm{\&theta;}-\frac{2\mathrm{\&pi;}}{3}))$

, can be similar to than not being under the very big situation (the salient pole ratio is 1.2～1.3) at salient pole:

So have: ${\mathrm{\&Delta;v}}_{\mathrm{sy}}=\frac{V_{\mathrm{dc}}\left(\frac{\sqrt{3}}{2}{\mathrm{<figure-callout\; id="2"\; label="L\; g"\; filenames="S2008100694727E00032.png"\; state="\{\{state\}\}">L</figure-callout>}}_g\cos (2\mathrm{\&theta;}+\frac{5\mathrm{\&pi;}}{6})\right)}{\frac{L_d+{\mathrm{<figure-callout\; id="2"\; label="L\; q"\; filenames="S2008100694727E00032.png"\; state="\{\{state\}\}">L</figure-callout>}}_q}{2}}$

Δv _cs(T)＝v _cs(t1)-v _cs′(t2)＝L _ac[i _a(t1)-i _a′(t2)]+L _bc[i _b(t1)-i _b′(t2)]＝0

$\mathrm{\&Delta;}{v}_{\mathrm{cy}}\left(T\right)=v_{\mathrm{cy}}\left(t1\right)-{v^{\&prime;}}_{\mathrm{cy}}\left(t2\right)={\mathrm{\&Delta;v}}_{\mathrm{sy}}\left(T\right)+{\mathrm{\&Delta;v}}_{\mathrm{cs}}\left(T\right)=\frac{V_{\mathrm{dc}}\left(\frac{\sqrt{3}}{2}{\mathrm{<figure-callout\; id="2"\; label="L\; g"\; filenames="S2008100694727E00032.png"\; state="\{\{state\}\}">L</figure-callout>}}_g\cos (2\mathrm{\&theta;}+\frac{5\mathrm{\&pi;}}{6})\right)}{\frac{L_d+L_q}{2}}---\left(35\right)$

According to formula (35) as can be known, at a mutually and during b switches on mutually, as if Δ v _Cy(T)=0, can judge c phase back electromotive force zero-crossing, the 30 ° of electrical degrees that lag behind again are the commutation moment next time.Because this method is to judge the zero crossing of back-emf by the inductance of indirect detection motor stator winding, therefore irrelevant with rotating speed of motor with the relating to parameters of motor self, so among the present invention this method is referred to as indirect inductance method.

The invention has the beneficial effects as follows: overcome traditional " back electromotive force method " (is lower than rated speed 20%) and lost efficacy when motor low speed problem, can greatly enlarge the working range of position Sensorless Control, when utmost point low speed (rated speed 5%), also can work reliably by method of the present invention.

Description of drawings

Fig. 1 is the main circuit of permanent-magnet brushless DC electric machine

Fig. 2 is the back emf waveform and the corresponding conducting phase of brshless DC motor

Fig. 3 is a bipolarity PWM copped wave mode of operation

Fig. 4 is the a+b-conductive path

Fig. 5 is the a-b+ conductive path

Fig. 6 is an a+b-conducting mode of operation

Fig. 7 is an a-b+ conducting mode of operation

Fig. 8 is the current waveform in a typical PWM cycle

Fig. 9 is the hardware control system block diagram

Figure 10 is a rotor-position pre-determined bit schematic diagram

Figure 11 is a main program flow chart

Figure 12 is medium event handling subprogram

Figure 13 is slow event handling subprogram

Figure 14 is twice locator program

Figure 15 is the PWM interruption subroutine

Figure 16 is the AD interrupt service subroutine

Figure 17 is the TIMER2 interruption subroutine

Embodiment

Below in conjunction with accompanying drawing enforcement of the present invention is elaborated.

The hardware components explanation

As shown in Figure 9, a kind of built-in permanent magnetic brushless DC motor control system of position-sensor-free is made of microprocessor controller 1, IPM module and driver element 2, current detection circuit 3, phase voltage testing circuit 4, permanent-magnet brushless DC electric machine 7, driving and the control power supply 8 of exporting display circuit 5, control input circuit 6, embedded rotor structure and fault detect and protective circuit 9; Microprocessor controller 1 calculates the position signalling of rotor according to detected voltage, current signal, produces to comprise six road pwm control signals; After pwm control signal process IPM module and driver element 2 conversion, produce the high-voltage signal that width changes with PWM, drive IPM module and driver element 2 circulation conductings, make motor rotation.The control method of described microprocessor controller 1 is:

(1) adopts the method for " continuous quadratic " location to give the rotor pre-determined bit, determined the initial position of rotor;

(2) t1 in same PWM period T and the t2 no power voltage to earth mutually of sampling constantly, and calculate this difference of twice is if the inferior difference of N continuous (3～5) is then assert the back electromotive force zero-crossing of this phase less than set point; According to the 60 ° of electrical degrees of being separated by of twice adjacent back electromotive force zero-crossing in the last circulation, thereby calculate 30 ° of electrical degrees, the commutation that the 30 ° of electrical degrees that lag behind again on the basis of back-emf zero crossing are next phase constantly;

(3) motor has whenever circled 6 commutations constantly, just can calculate the mean angular velocity during the commutation interval 2 times as long as record 2 commutations time interval constantly; The set-point of electric current loop carries out the PI computing by the rotating speed deviation en to der Geschwindigkeitkreis and obtains, and the deviate ei of electric current reconciles computing by current PI and becomes PWM ripple duty ratio, thus the control motor speed.

Wherein, microcontroller of the present invention is a core with the dsPIC30F6010 of Microchip company, electric current, voltage signal are gone into microcontroller dsPIC30F6010 and are handled through the A/D sampling is laggard, produce the various control signals that comprise six road PWM that drive dc brushless motor then, finish control system.Introduce its major part function below:

1. microprocessor controller 1: microprocessor is a core with the dsPIC30F6010 of Microchip company, calculates the position signalling of rotor according to detected voltage, current signal, produces to comprise six road pwm control signals, and these functions are mainly realized by software.

2.IPM module and driver element 2: this control circuit at first inputs to power driving device with the control signal that microprocessor control circuit produces, after the power driving device conversion, produce the high-voltage signal that width changes with PWM, drive IPM module circulation conducting, make motor rotation.

3. current detection circuit 3: use Hall current sensor or realize the detection of electric current by a low resistance.

4. the phase voltage testing circuit 4: working voltage Hall element or realize the detection of phase voltage by an electric resistance partial pressure.

5. the shows signal circuit 5: be made of with relevant drive circuit charactron, be used for showing the various states and the failure code digital information of work system.

6. control input circuit 6: importing given can be also can select velocity setting to realize by a potentiometer by serial communication from host computer.

7. drive and control power supply 8: provide power supply to associated components.

8. fault detect and protective circuit 9: detect overcurrent, overvoltage, IPM module excess temperature, circuit sends alarm signal, and control circuit sends signal and blocks the power drive control circuit, and its output is connected with the signal input port of microprocessor control circuit.

System comprises also-fault detect and protective circuit (9) that detect overcurrent, overvoltage, IPM module excess temperature, circuit sends alarm signal, and control circuit sends signal and blocks the power drive control circuit;

Device is selected and relevant issues in the hardware designs

1.IPM the selection of module: the IPM module that native system adopts Mitsubishi to produce, this inside modules is integrated 6 high-power IGBTs connect into the three phase full bridge circuit, and comprise a temperature detection thermistor, and are very easy to use.Important parameters has: the collector electrode of each IGBT-emitter maximum is born voltage, the maximum collector current under 75 ℃, and peak current between the conducting time-delay, between turn off delay time etc., to satisfy the requirement of design system.

2.PWM the isolation design of signal and rub-out signal: the pwm signal of chip output is the driving power switching device directly, therefore use high-speed light to make the isolated component of control signal every HCPL-4504, the HCPL-4504 delay time is extremely short, add that every output signal end 0.1uF electric capacity is as decoupling capacitor in high-speed light, outer meeting resistance is as pull-up resistor, and the big more then pull-up resistor of the IPM modular power of control is more little; In addition, the instantaneous common mode of HCPL-4504 is 15kV/us, and these all help power device and work reliably.The output of fault-signal only needs common light every getting final product, and for example low speed light is every PC817, and is low because low speed light is divided into, the current transfer ratio height, and this makes it be very suitable for the lower switching signal of transmission frequency.

3. internal system power supply design: dual mode is adopted in power supply to CPU, and a kind of is the 5V power supply signal of the socket type of employing standard, and another kind is the 5V power supply signal that adopts the input of plug formula terminal,

4.A/D the design of sample circuit: need to gather voltage signal, current signal etc. in native system, voltage signal is divided into three phase terminals voltage and busbar voltage again, so the photoelectric signal collection just needs 4 A/D passages in design.During measurement voltage signal, because three-phase voltage and busbar voltage are all bigger, so adopted voltage signal is dropped to after the electric resistance partial pressure-5V is between+the 5V, deliver to follow circuit then and enter voltage lifting and adjustment, and carry out filtering, the signal that ensures to CPU is the burning voltage signal of 0V-5V.Current sensor LA108-P is adopted in the measurement of electric current, and its transfer ratio is 1: 2000, big electric current can be become little current signal, obtains magnitude of voltage by series resistor then, thereby calculates the current value that is collected by magnitude of voltage.

5. charactron display circuit design: display unit can be used for showing motor speed, input and output voltage amplitude and system failure code digital information.The SPI module of dsPIC30F6010 chip and shift register chip 74HC595 realize the static demonstration of charactron.74HC595 is one 8 a shift register, and it can convert 8 bit data of serial input and line output to.Its inside comprises that 8 bit strings go into and go out shift register and 8 ternary output latches.The clock input SRCLK of register links to each other with the SCK pin of dsPIC30F601, and latch signal RCLK links to each other with RG9.When the level of SRCLK jumps to when high from low, the data of register are inserted latch, and the latch dateout is to charactron.

The software section explanation

Control Software is the core of system, mainly finishes following function:

1. what system adopted is method for controlling position-less sensor, must adopt the method for software to determine the initial position of rotor.The initial position of rotor has determined the conducting order of inverter, but motor position of rotor before startup is unknown, therefore at first will give the rotor pre-determined bit.When system powered on beginning, given one group of trigger impulse formed a constant amplitude, the constant magnetic flux of direction in air gap arbitrarily, as long as guarantee that its amplitude is enough big, this magnetic flux just can be positioned rotor on this direction within a certain period of time by force so.The position fixing process signal of Figure 10 (a) expression rotor when the 1st pipe and the 6th pipe conducting.Location rear motor rotor d axle overlaps with the stator winding flow direction. so just determined the initial position of rotor.But because the uncertainty of motor rotor position when static, if its position is in the position of Figure 10 (b) just before the location, behind conducting the 1st pipe and the 6th pipe, stator winding resultant flux FI and rotor d axle clamp angle are 180 °, this moment, rotor can not rotate to the position shown in Figure 10 (a), the location failure.In order to address this problem, used " continuous quadratic " localization method, its positioning principle is shown in Figure 10 (c).Promptly on the basis of locating for the first time, follow the next sector of conducting, no matter locate successfully or fail the first time like this, it must be correct locating for the second time.

2. the three-phase brushless dc motor revolution just needs commutation once for 60 °, and whenever going around needs commutation six times, therefore needs six commutation signals.As can be known permanent-magnet brushless DC electric machine is carried out bipolarity copped wave control according to principle of the present invention, as shown in Figure 8, a PWM in the cycle t1 and t2 (t1 gets 0.5T ₀, t2 gets 0.5 (T+T ₀), wherein T is the cycle of PWM, T ₀=ρ T, ρ is a duty ratio) constantly sample no power relatively electromotive force and calculate their electrical potential difference, if the inferior electrical potential difference of N continuous (3～5) is less than set point (very little numerical value, this numerical value should be set according to real system), then can think the back-emf zero crossing of no power phase, and then it is postponed 30 ° of electrical degrees be next commutation constantly.T1 sampling is constantly finished in ADC interrupts, and finishes t2 sampling constantly in TIMER2 interrupts, and carries out twice difference calculating, if the inferior difference of N continuous (3～5) less than set point, then shows the back electromotive force zero-crossing of no power phase.Twice adjacent back electromotive force zero-crossing, the 60 ° of electrical degrees of being separated by, thus calculate 30 ° of electrical degrees.The 30 ° of electrical degrees that lag behind again on the basis of back-emf zero crossing are the commutation moment of next phase.

3. set up the brushless DC motor without position sensor governing system of two closed loops of rotating speed, electric current.Motor has whenever circled 6 commutations constantly, just can calculate the mean angular velocity during the commutation interval 2 times as long as record 2 commutations time interval constantly; The set-point of electric current loop carries out the PI computing by the rotating speed deviation en to der Geschwindigkeitkreis and obtains, and the deviate ei of electric current reconciles computing by current PI and becomes PWM ripple duty ratio, thus the control motor speed.

4. accept outside rotational speed setup signal, the control motor speed.

5. show some parameters such as motor speed, duty ratio in real time by liquid crystal and charactron.

The design of control system main program:

After starting sequence was finished, the control of CPU was given main program, carried out various initialize routines and System self-test that the user writes, mainly was each device register of initialization and system global variables, finished the I/O configuring ports, the operating state of check system.After initialization is finished, if system all show then normally and normally treat machine information that the subprogram of calling twice location is determined the initial position of rotor, enters the main program circulation.

Main program calls medium button.onrelease and slow event handling procedure; Medium button.onrelease is realized calculating and adjusting spinner velocity, and function such as voltage control, and every 10ms carries out once; The slow event handling procedure is realized functions such as user interface demonstration and keyboard operation, and every 100ms carries out once.In order to prevent program fleet, used WatchDog Timer WDT.Because main program is a big circulation, so clear house dog WDT is wanted in each circulation.Main program flow chart as shown in figure 11, medium button.onrelease as shown in figure 12, slow button.onrelease as shown in figure 13, the subprogram of twice location is as shown in figure 14.

Interrupt service routine mainly comprises:

Core Feature of the present invention-" inductance method indirectly " realizes that in interrupt service routine it mainly is made up of PWM interrupt service routine, ADC interrupt service routine and TIMER2 interrupt service routine.

PWM interrupt service routine: be used for determining the state variable of medium button.onrelease and full gear button.onrelease, can be triggered on time, use PWM to interrupt starting ADC constantly and interrupt at t1 to guarantee these two kinds of handling procedures.The subprogram flow graph as shown in figure 15.

ADC interrupt service subroutine: be used to measure the numerical value of CH0 (VPH), CH1 (VDC), CH2 (IBUS) and CH3 (POT), and judge whether overvoltage or overcurrent, obtain the position then and check zero passage.DsPIC30F6010 has 16 tunnel 10 high-speed a/d ALT-CH alternate channel, simultaneously four paths are sampled, needing the signal of sampling simultaneously is busbar voltage-VDC, bus current-IBUS, every phase terminal voltage-VPH, adjustable potentiometer voltage-POT four tunnel, is that four road AD sampling channels are sampled simultaneously so need to set the A/D module.Because four groups of A/D signals sample simultaneously, so need configuration ADCHS input mask register to guarantee that four groups of signals sample simultaneously, wherein three phase voltage signals just can only measure a phase voltage value sometime by the scan mode input.By configuration ADCHS, thereby select AN3, AN4, AN5, as three constant inputs, corresponding S/H amplifier is respectively CH1, CH2, CH3.AN6, AN7, AN8 is the scanning input, corresponding S/H amplifier is CH0.The programming flowchart of A/D sampling as shown in figure 16.

TIMER2 interrupt service subroutine: start TIMER2 constantly at the t2 in each cycle and interrupt, check the voltage to earth of no power phase, and the state of sensor-less operation is handled.In invention, use indirect inductance method to detect the zero crossing of back electromotive force, the sampling channel of hardware designs is sampled to the terminal voltage of no power phase, then through software algorithm, draws the commutation information of rotor by terminal voltage information, detects back electromotive force zero-crossing.Twice adjacent back electromotive force zero-crossing time difference is the required time that current rotor turns over 60 ° of electrical degrees, and rotor has been null event constantly from commutation and has been differed 30 ° of electrical degrees, so after detecting null event, only needing 30 ° of electrical degrees of compensation is exactly the commutation moment, in the design of native system, from QEI time for reading mark, then 30 ° of electrical degrees of compensation.The subprogram flow graph as shown in figure 17.

Claims (2)

1, a kind of built-in permanent magnetic brushless DC motor control system of position-sensor-free comprises the permanent-magnet brushless DC electric machine (7) of microprocessor controller (1), IPM module and driver element (2) and embedded rotor structure; It is characterized in that: microprocessor controller (1) calculates the position signalling of rotor according to detected voltage, current signal, produces to comprise six road pwm control signals; After pwm control signal process IPM module and driver element (2) conversion, produce the high-voltage signal that width changes with PWM, drive IPM module circulation conducting, make permanent-magnet brushless DC electric machine (7) running of embedded rotor structure; Described microprocessor controller (1) control method is:

(1) adopts the rotor pre-determined bit of the method for " continuous quadratic " location, determine the initial position of permanent-magnet brushless DC electric machine (7) rotor to permanent-magnet brushless DC electric machine (7);

(2) t1 in same PWM period T and the t2 no power voltage to earth mutually of sampling constantly, and calculate this difference of twice is if the inferior difference of N continuous (3～5) is then assert the back electromotive force zero-crossing of this phase less than set point; According to the 60 ° of electrical degrees of being separated by of twice adjacent back electromotive force zero-crossing in the last circulation, thereby calculate 30 ° of electrical degrees, the commutation that the 30 ° of electrical degrees that lag behind again on the basis of back-emf zero crossing are next phase constantly;

(3) described permanent-magnet brushless DC electric machine (7) has whenever circled 6 commutations constantly, just can calculate the mean angular velocity during the commutation interval 2 times as long as record 2 commutations time interval constantly; The set-point of electric current loop carries out the PI computing by the rotating speed deviation en to der Geschwindigkeitkreis and obtains, and the deviate ei of electric current reconciles computing by current PI and becomes PWM ripple duty ratio, thus the rotating speed of control permanent-magnet brushless DC electric machine (7).

2, the built-in permanent magnetic brushless DC motor control system of position-sensor-free according to claim 1, it is characterized in that: system also comprises-fault detect and protective circuit (9), detect overcurrent, overvoltage, IPM module excess temperature, circuit sends alarm signal, and control circuit sends signal and blocks the power drive control circuit; Its output is connected with the signal input port of microprocessor control circuit.

Images