CN1825752A - Sliding moding structure direct torque servo-driver - Google Patents

Abstract

This invention relates to a direct torque servo drive device of a slip form inversion structure including a commutation inversion output circuit, a control circuit and a control object, in which, the commutation output circuit includes a commutation filter unit and an IPM inversion unit, the control circuit includes a DSP processor, an IPM isolation drive and protection circuit, a current sample circuit and a voltage sample circuit, the control object is a three phase permanent magnetic synchronous motor controlled by a control program in the DSP processor of the inserted control circuit.

Description

A kind of sliding moding structure direct torque servo-driver

Technical field

The invention belongs to the electric drive technology field, particularly a kind of sliding moding structure direct torque servo-driver.

Background technology

Current, the development of international exchange speed adjusting technique is quite rapid, and the frequency converter of various application of advanced control strategies emerges in an endless stream.By contrast, the ac speed control technology of China is started late, though development in recent years is very fast, still has no small gap with the development level of western developed country.Therefore along with development of modern industry, more and more higher to the requirement of ac speed control technology, developing high performance AC speed regulator is a urgent task, and the development of Chinese national economy is had great importance.Direct torque control realizes simple, good dynamic characteristics and the new technology that advantages such as parameter of electric machine reliability is little is become a kind of tool potentiality with its control algolithm.The high-grade frequency converter of asynchronous machine of nineteen ninety-five Sweden ABB AB first employing direct torque control scheme just emerges, and they think that direct torque control will be the classic control mode of alternating current machine of future generation.In actual applications, the installation rate transducer had both increased installation cost, reduced device reliability again, so direct torque control Speedless sensor technology becomes a popular direction, but magnetic linkage and torque pulsation are big during direct torque control low speed, and this is the major defect that hinders its development.

Summary of the invention

Problem at prior art exists the invention provides a kind of sliding moding structure direct torque servo-driver based on space vector modulation technique.

The structure of apparatus of the present invention comprises commutation inversion output circuit, control circuit and three parts of controlling object as shown in Figure 1.The commutation inversion output circuit comprises rectification filtering unit and IPM inversion unit.Control circuit comprises dsp processor, IPM isolation drive and protective circuit, current sampling circuit, voltage sampling circuit.Controlling object is a three-phase permanent-magnetic synchronous motors.The input of IPM inversion unit main power source P, N end links to each other with rectification circuit output, voltage sampling circuit is gathered voltage difference between P, N, IPM lead-out terminal U, V, W link to each other with permagnetic synchronous motor, and V, W link to each other with the two-way current sampling circuit by two Hall current sensors again.16 road control terminals of IPM link to each other with IPM isolation drive protective circuit, and the input of IPM isolation drive and the output of protective circuit and electric current, voltage sampling circuit output link to each other with dsp processor.

The commutation inversion output circuit as shown in Figure 2, isolating transformer is connected with voltage regulator after two contact K are connected to rectifier bridge, the anode of diode is connected to the N end of IPM main power source in the rectifier bridge, the negative electrode of diode is connected to the P end of IPM main power source, the three-phase current of IPM output is by lead-out terminal U, V, W is connected to threephase motor.P, N are the IPM main power source input terminal behind the rectifying conversion smothing filtering of frequency converter, and P is an anode, and N is a negative terminal, and B is the braking lead-out terminal.Wherein isolating transformer plays the effect of protection; Rectification filtering unit adopts the uncontrollable rectifier system of bridge-type, and big capacitor filtering can obtain to be suitable for the constant voltage that IPM works like this.The input of inversion unit IPM main power source (P, N), braking output (B), (W), main terminal M5 screw can be realized current delivery to lead-out terminal for U, V.

The core of control circuit is the TMS320LF2407 processor, adopt LF2407 assessment version (LF2407EVM plate), its peripheral circuit configuration block diagram such as Fig. 3, the main interface of EVM plate comprises traget ROM, analog interface, CAN interface, serial boot ROM, user lamp and switch, RS232 interface, SPI data-interface and expansion interface.The necessary part of this device comprises that also power supply, crystal oscillator, jtag interface, 128K word length do not have the static memory of delay, simulation extends out interface, PWM extends out interface.

The circuit connecting relation of the LF2407 evaluation board several sections of using in this drive unit is: the address bus of TMS320LF2407 meets enlarging P3 outside the address bus of static memory U3, U4 and the address respectively.The data/address bus of TMS320LF2407 meets enlarging P3 outside the data/address bus of U3, U4 and the address respectively.The read-write enable pin of TMS320LF2407 connects 17,41 pins of U3, U4 respectively.The program space gating pin of TMS320LF2407 connects 6 pins of U3, and the data space gating pin of TMS320LF2407 connects 6 pins of U4.The JTAG pin of TMS320LF2407 meets P5, and P5 links to each other with an end of simulator, and the other end links to each other with PC by LPT.The analog-to-digital conversion pin of TMS320LF2407 connects 23,24,5,6,7,8,9,10,11,12,13,14,15,16,19,20 pins of the outer enlarging P2 of simulation respectively.The PWM interface of TMS320LF2407 task manager diffuses into 3,4,5,6,7,8,12,13,14,9,10,11,21,22,24 pins of P1,20,25,26,27,29 pins of P4 outward.1 pin of the external 15M crystal oscillator of the 123 pins U22 of TMS320LF2407.The analog references power pin 116,117 of TMS320LF2407 connects 4 and 11 pins of U19 respectively.The digital reference power pin 29,50,86,129,4,42,67,77,95,141 of TMS320LF2407 connects 17,18 and 19 pins of 3.3V voltage source module U12.It digitally is 9 and 10 pins of U12 that 28,49,85,128,3,41,66,76,94,125,140 pins of TMS320LF2407 connect.

IPM isolated drive circuit shown in Fig. 5 (a); 1,3,4,5,6,7 of its input PORT6 receives 3,4,5,6,7,8 pins of assessment version P1, and 7 pins that the fault output signal of IPM is received TMS320LF2407 by optocoupler in the IPM protective circuit shown in Fig. 5 (b) extend out 26 pins into P1.

Fig. 6 is the current sampling circuit schematic diagram, it is input as the output signal of two-phase Hall current sensor, Hall current sensor is powered by positive and negative 15 DC voltage-stabilizing electric current, its input is that any two phase conductors with the U of motor, V, W pass around two circles from the center of Hall current sensor, corresponding output will induce the current signal of 1000:2, be defined as A mutually with B mutually, as the input of current sampling circuit, A, B phase current sampling circuit are identical.Current sampling circuit is by the adjusting to adjustable resistance 503, adjustable resistance 202, signal is adjusted between 0～3.3V, it is sent into the AD conversion pin of DSP again, select 23 and 24 pins of P2 on the EVM at this, wherein amplifier is at the indirect decoupling capacitor of voltage and ground.

Voltage sampling circuit, P, the N of Hall voltage transducer master edge joint IPM end, voltage sensor output signal M connects the AD conversion pin of DSP through processing of circuit output signal as shown in Figure 7, selects 5 pins of P2 on the EVM at this.

The control of apparatus of the present invention realizes that by the control program that embeds in the control circuit dsp processor its control procedure is carried out (as shown in Figure 8) according to the following steps:

Step 1, initialization;

Step 2, rotor initial alignment;

Step 3, permission INT1, INT2 interrupts;

Step 4, startup T1 underflow are interrupted;

Step 5, interrupt latency;

Step 6, T1 Interrupt Process;

Step 7, protection Interrupt Process;

Step 8, end.

Wherein protection Interrupt Process process is carried out (as shown in Figure 9) according to the following steps in the step 7:

Step 1 is forbidden all interruptions;

Step 2 is blocked PIM;

Step 3 interrupts returning.

T1 Interrupt Process process is carried out (as shown in figure 10) according to the following steps in the step 6:

Step 1 keeps the scene intact;

Step 2 current-voltage sampling;

Step 3 current and voltage data is handled;

Step 4 magnetic linkage is estimated;

Step 5 torque is estimated;

Whether speed regulation of step 6 is to enter step 7, otherwise enters step 9;

Step 7 velocity estimation;

Step 8 speed PI regulates;

Step 9 magnetic linkage and torque Sliding mode variable structure control device are regulated;

Step 10 space vector modulation;

It is on-the-spot that step 11 is recovered;

Step 12 interrupts returning.

T1 underflow interrupt procedure is mainly finished Speedless sensor sliding moding structure direct torque control calculating in step 6.

Enter and at first keep the scene intact after interrupting, start the AD conversion then, the electric current of being sent back by peripheral circuit, magnitude of voltage are collected in the middle of the DSP.Gathering the data of returning at first is in the middle of the result register (RESUTLx) that is stored in separately, from result register read current, magnitude of voltage, electric current is carried out coordinate transform obtain biphase current value under the rest frame.Enter for the first time when interrupting, because rotor has carried out initial alignment, so can obtain the value of two phase voltages under the rest frame first time.The value that has obtained current/voltage just can be carried out the estimation of magnetic linkage, and then carries out torque and estimate.The requirement of response speed is far from magnetic linkage to speed ring and the torque ring is high like that to the requirement of response speed, stipulates that therefore the T1 underflow interrupts all will carrying out the adjusting of magnetic linkage and torque each time, and interrupt just carrying out the estimation and the PI adjusting of a speed per 20 times.Therefore before admission velocity estimation and PI adjusting, to judge whether to want admission velocity to regulate once.Velocity estimation adopts following algorithm to realize:

${\mathrm{\ω}}_r(k+1)={\mathrm{\ω}}_s(k+1)=\frac{(V_{\mathrm{\β}}\left(k\right)-i_{\mathrm{\β}}\left(k\right)){\mathrm{\ψ}}_{\mathrm{\α}}\left(k\right)-(V_{\mathrm{\α}}\left(k\right)-i_{\mathrm{\α}}(k\left)\right){\mathrm{\ψ}}_{\mathrm{\β}}(k)}{{\mathrm{\ψ}}_{\mathrm{\α}}{\left(k\right)}^2+{\mathrm{\ψ}}_{\mathrm{\β}}{\left(k\right)}^2}$

Subscript α in the formula, β represent static two coordinate systems;

ω _r(k+1)---the k+1 time spinner velocity estimated value;

ω _s(k+1)---the k+1 time stator field rotary speed estimated value;

V---voltage; I---electric current; ψ---stator magnetic linkage;

K---the k time sampling or result of calculation.

Obtained the velocity estimation value, obtained velocity error doing difference, carried out speed PI and regulate, the set-point of output torque with the velocity setting value.

After the velocity estimation value obtains the position of rotor is estimated can realize with following formula:

θ _r(k+1)＝θ _r(k)+ω _r(k)ΔT

By the DC bus-bar voltage that AD samples and, looking into rotor-position estimation θ _rThe sine and cosine table, obtain (α, β) V under the coordinate system _α(k+1) and V _β(k+1) value is to be used for the velocity estimation of rotor next time.

Calculate the set-point and the estimated value of magnetic linkage torque according to the current value of given galvanic current value and current sampling circuit sampling, use the Sliding mode variable structure control device that these two amounts are regulated.

The input of Sliding mode variable structure control device is respectively the error of magnetic linkage and torque, is output as α under the rest frame, the component of voltage of β axle.

${\underset{\‾}{V}}_{\mathrm{ref}}=\left[\begin{array}{c}{V}_{\mathrm{\α}}(k+1)\\ V_{\mathrm{\β}}(k+1)\end{array}\right]=-{\underset{\‾}{D}}^{-1}\left(k\right)\left[\begin{array}{cc}{\mathrm{\μ}}_1& 0\\ 0& {\mathrm{\μ}}_2\end{array}\right]\left[\begin{array}{c}\mathrm{sign}\left(S_1\right)\\ \mathrm{sign}\left(S_2\right)\end{array}\right]$

Wherein ${\underset{\‾}{D}}^{-1}\left(k\right)=\frac{1}{\left|D\right|}\left[\begin{array}{cc}-{2\mathrm{\ψ}}_{\mathrm{\β}}\left(k\right)& \frac{3}{2}p(\frac{{\mathrm{\ψ}}_{\mathrm{\α}}\left(k\right)}{L_s}-i_{\mathrm{\α}}\left(k\right))\\ {2\mathrm{\ψ}}_{\mathrm{\α}}\left(k\right)& -\frac{3}{2}p(i_{\mathrm{\β}}\left(k\right)-\frac{{\mathrm{\ψ}}_{\mathrm{\β}}\left(k\right)}{L_s})\end{array}\right]$

$\left|D\right|=3p[({\mathrm{\ψ}}_{\mathrm{\α}}\left(k\right)i_{\mathrm{\α}}\left(k\right)+{\mathrm{\ψ}}_{\mathrm{\β}}\left(k\right)i_{\mathrm{\β}}\left(k\right))-\frac{1}{L_s}(({\mathrm{\ψ}}_{\mathrm{\α}}{\left(k\right)}^2+{\mathrm{\ψ}}_{\mathrm{\β}}{\left(k\right)}^2)]$

$\mathrm{sign}\left(S_i\right)=\left\{\begin{array}{c}1,S_i>{\mathrm{\&lambda;}}_i\\ -1,S_i<{-\mathrm{\&lambda;}}_i\\ \frac{S_i}{{\mathrm{\&lambda;}}_i},\left|S_i\right|<{\mathrm{\&lambda;}}_i\end{array}\right.$

$\left\{\begin{array}{c}{\mathrm{<figure-callout\; id="1"\; label="S"\; filenames="A20061004584900161.png,A20061004584900171.png"\; state="\{\{state\}\}">S</figure-callout>}}_1=e_T\left(k\right)+{\mathrm{<figure-callout\; id="1"\; label="K"\; filenames="A20061004584900161.png,A20061004584900171.png"\; state="\{\{state\}\}">K</figure-callout>}}_1\underset{i=0}{\overset{k}{\mathrm{\&Sigma;}}}{e}_T\left(i\right)-e_T\left(0\right)\\ {\mathrm{<figure-callout\; id="2"\; label="S"\; filenames="A20061004584900161.png,A20061004584900171.png"\; state="\{\{state\}\}">S</figure-callout>}}_2=e_{\mathrm{\&psi;}}\left(k\right)+{\mathrm{<figure-callout\; id="2"\; label="K"\; filenames="A20061004584900161.png,A20061004584900171.png"\; state="\{\{state\}\}">K</figure-callout>}}_2\underset{i=0}{\overset{k}{\mathrm{\&Sigma;}}}{e}_{\mathrm{\&psi;}}\left(i\right)-e_{\mathrm{\&psi;}}\left(0\right)\end{array}\right.$

P in the formula---motor number of pole-pairs; L _s---the stator winding inductance; μ ₁, μ ₂---the sliding formwork coefficient;

I=1,2, λ _i---the width of sliding formwork BL (Boudary layer);

K ₁, K ₂---the sliding formwork gain.

Voltage given value under the rest frame that obtains is the input of space vector modulation (SVPWM), utilize space vector modulation (SVPWM) technology to generate the action time of adjacent two voltage vectors again, change into corresponding value and write three comparand registers (CMPRx), export corresponding pwm signal and drive IPM, finally carry out the SERVO CONTROL of motor.

Apparatus of the present invention have been saved velocity transducer, have saved cost; Adopt sliding mode control algorithm to improve the antijamming capability of control system, overcome owing to various external disturbances cause the control system instability; This device control precision height can be applicable to the occasion that speed adjusting performance is had relatively high expectations.

Description of drawings

Fig. 1 is apparatus of the present invention structure chart;

Fig. 2 is apparatus of the present invention commutation inversion output circuit schematic diagram;

Fig. 3 is dsp processor and a peripheral circuit configuration block diagram in apparatus of the present invention;

Fig. 4 (1) is dsp processor and part peripheral circuit schematic diagram thereof, and (2) are the peripheral crystal oscillating circuit schematic diagram of DSP,

(3) be the circuit theory diagrams of P5,

(4) be voltage module TPS73HD318 and peripheral circuit schematic diagram thereof,

(5) be the annexation figure of enlarging P3 outside the address, (6) are the annexation figure of control mouthful P4,

(7) be the annexation figure of the outer enlarging P2 of simulation, (8) are the annexation figure that extends out I/O mouth P1,

(9) be RS232 and peripheral circuit schematic diagram thereof, (10) are the annexation figure of static memory U3,

(11) be the annexation figure of static memory U4, (12) are the annexation figure of U9 (74LS245),

(13) be the annexation figure of U2 (74LS245),

(14) be U19, U16, U10, U14, U8, U1 and peripheral circuit schematic diagram thereof,

(15) be the annexation figure of U17,

(16) be U20 (TLC2274) and peripheral circuit schematic diagram thereof, (17) are the peripheral circuit schematic diagram of U13C, B,

(18) be U13F peripheral circuit schematic diagram, (19) are U7 and peripheral circuit schematic diagram thereof,

(20) be U20 and peripheral circuit schematic diagram thereof,

(21) be the peripheral circuit schematic diagram of JP16, JP6,

(22) the peripheral circuit schematic diagram of JP9, (23) are the peripheral circuit schematic diagram of JP5,

(24) the peripheral circuit schematic diagram of JP8,

(25) be the peripheral circuit schematic diagram of J1, J2, JP7 and switch SW 3,

(26) be the annexation figure of U11 (74LS245), (27) are U6, U5 and peripheral circuit schematic diagram thereof,

(28) for to be that the annexation figure of L7, C28, (29) are the circuit theory diagrams of 5VO,

(30) be the annexation figure of L3,

(31) be the annexation figure of L6, C7, C44, C55, C52, C53C51,

(32) be the annexation figure of L1, C2, C13, C27, C49;

Fig. 5 (a) is an IPM isolated drive circuit schematic diagram, (b) is IPM protective circuit schematic diagram;

Fig. 6 (a) is A phase current sampling circuit theory diagrams, (b) is B phase current sampling circuit theory diagrams;

Fig. 7 is apparatus of the present invention voltage sampling circuit schematic diagram;

Fig. 8 is a control program FB(flow block) in the dsp processor of apparatus of the present invention;

Fig. 9 is the protection Interrupt Process process flow block diagram of apparatus of the present invention;

Figure 10 is the T1 Interrupt Process process flow block diagram of apparatus of the present invention.

Embodiment

The structured flowchart of apparatus of the present invention comprises commutation inversion output circuit, control circuit and three parts of controlling object as shown in Figure 1.The commutation inversion output circuit comprises rectification filtering unit and IPM inversion unit.Control circuit comprises dsp processor, IPM isolation drive and protective circuit, current sampling circuit, voltage sampling circuit.Controlling object is a three-phase permanent-magnetic synchronous motors.The input of IPM inversion unit main power source P, N end links to each other with rectification circuit output, voltage sampling circuit is gathered voltage difference between P, N, IPM lead-out terminal U, V, W link to each other with permagnetic synchronous motor, and V, W link to each other with the two-way current sampling circuit by two Hall current sensors again.16 road control terminals of IPM link to each other with IPM isolation drive protective circuit, and the input of IPM isolation drive and the output of protective circuit and electric current, voltage sampling circuit output link to each other with dsp processor.

The commutation inversion output circuit as shown in Figure 2, isolating transformer is connected with voltage regulator after two contact K are connected to rectifier bridge, the anode of diode is connected to the N end of IPM main power source in the rectifier bridge, the negative electrode of diode is connected to the P end of IPM main power source, the three-phase current of IPM output is by lead-out terminal U, V, W is connected to threephase motor.P, N are the main power source input terminal behind the rectifying conversion smothing filtering of frequency converter, and P is an anode, and N is a negative terminal, and B is the braking lead-out terminal.Wherein isolating transformer plays voltage stabilizing, prevents that the electrical network high-frequency signal from disturbing the effect of commutation inversion output circuit; Rectification filtering unit adopts the uncontrollable rectifier system of bridge-type, and big capacitor filtering can obtain to be suitable for the constant voltage that IPM works like this.The input of inversion unit IPM main power source (P, N), braking output (B), (W), main terminal M5 screw can be realized current delivery to lead-out terminal for U, V.

The core of control circuit is a dsp processor, adopt the EVM plate to carry out full speed and checks, comprise on the plate of data storage, 128K word length on the sheet of 544 word lengths on memory, the sheet UART and a MP7680D/A transducer on read-only flash memory, the sheet the LF2407 code.The device of installing on the plate can make evaluation board solve all kinds of problems.4 expansion interfaces can offer any evaluation circuits.Evaluation circuits is that the user develops voluntarily, can use a large amount of user interfaces and transform code, thereby shorten development time and cycle.

LF2407EVM has following feature:

(1) the LF2407 speed of service reaches 30MIPS and can realize that 128 word lengths do not have time-delay storage;

(2) 16 passages on the sheet, 10 analog to digital converters, ordering automatically;

(3) capture channel on multiple PWM of two task managers and the sheet;

(4) DAC7654 four-way digital to analog converter;

(5) UART on the sheet of band RS232 driver;

(6) the CAN interface of band driving;

(7) subscriber switch and light emitting diode lamp;

(8) 4 expansion interfaces (data address, I/O, control);

(9) provide IEEE1149 on the plate, 1 jtag interface for optionally assessing;

(10) 5V voltage input (voltage regulator on the 3.3V plate).

The block diagram of LF2407EVM basic configuration is seen Fig. 3, and the main interface of EVM plate comprises traget ROM, analog interface, CAN interface, serial boot ROM, user lamp and switch, RS232 interface, SPI data-interface and expansion interface.LF2407 is connected to the 128K word length does not have the delay static memory, and the I/O interface of an expansion is supported 65000 parallel I/O ports, and CAN interface and RS232 serial ports can be used as expansion interface on the sheet.The necessary part of this device comprises that power supply, crystal oscillator, jtag interface, 128K word length do not have the static memory of delay, simulation extends out interface, PWM extends out interface.

The DSP periphery is 127 of (as shown in Figure 4): TMS320LF2407 such as Fig. 4 (1) to this concrete connection of installing necessary partial circuit, 130,132,134,136,138,143,5,9,13,15,17,20,22,24,27 pins meet static memory U3 such as Fig. 4 (10) respectively, U4 such as Fig. 4 (11) (IS61LV6416) 7,8,9,10,13,14,15,16,29,30,31,32,35,36,37, outside 38 pins and the address 1 of enlarging P3 such as Fig. 4 (5), 2,3,4,5,6,7,8,9,10,11,12,13,14,15,16 pins.80,78,74,71,68,64,61,57,53,51,48,45,43,39,34,31 pins of TMS320LF2407 connect 19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34 pins of enlarging P3 outside 5,4,3,2,1,44,43,42,27,26,25,24,21,20,19,18 pins of U3, U4 and the address respectively.93,89 pins of TMS320LF2407 connect 17,41 pins of U3, U4.87 pins of TMS320LF2407 connect 6 pins of U3, and 84 pins of TMS320LF2407 connect 6 pins of U4.19,89,96,92,82,84,87 pins of TMS320LF2407 connect U17 such as Fig. 4 (15) 4,5,6,7,8,9,10,11 pins (GAL16V8) respectively.90,91,135,139,142,144 pins of TMS320LF2407 connect P5 such as Fig. 4 (3) 13,14,9 pins and 11,3,7,1 pin (JTAG) respectively.112,110,107,105,103,102,100,99,113,111,109,108,106,104,101,98 pins of TMS320LF2407 connect 23,24,5,6,7,8,9,10,11,12,13,14,15,16,19,20 pins of the outer enlarging P2 of simulation such as Fig. 4 (7) respectively.56,54,52,47,44,40,16,18,8,65,62,59,55,46,38,6 pins of TMS320LF2407 connect respectively extend out I 25,26,27,29 pins of 3,4,5,6,7,8,12,13,14,9,10,11 pins, P4 such as Fig. 4 (6) of O and PWM multiplex interface P1 such as Fig. 4 (8).83,79,88,81 pins of TMS320LF2407 connect extend out I O and 21,22,24 pins of PWM multiplex interface P1 and 20 pins of P4.The 123 pins external 15M crystal oscillator U22 of TMS320LF2407 such as 1 pin of Fig. 4 (2).The analog references power pin 116,117 of TMS320LF2407 connects U19 such as Fig. 4 (14) 4 and 11 pins (TLC2274) respectively.The digital reference power pin 29,50,86,129,4,42,67,77,95,141 of TMS320LF2407 connects 3.3V voltage source module U12 such as Fig. 4 (4) 17,18 and 19 pins (TOS73HD318).It digitally is 9 and 10 pins of U12 that 28,49,85,128,3,41,66,76,94,125,140 pins of TMS320LF2407 connect.25,19,26 pins of TMS320LF2407 connect 2 pins of 11,10 and the JP12 of RS232 interface U21 respectively, shown in Fig. 4 (9).72,70 pins of TMS320LF2407 connect 1 pin of U7 and 2 pins of JP2 respectively, shown in Fig. 4 (19).

Fig. 5 (a) is an IPM isolated drive circuit schematic diagram.1,3,4,5,6,7 of its input PORT6 receives 3～8 pins of EVM P1.Drive wiring and shielding and be the key that can the IPM module operate as normal.Design must be accomplished that driving power is isolated, wiring is reasonable, ground connection is correct, shielding is proper, connect decoupling capacitance simultaneously as far as possible, reduces parasitic capacitance, to reduce the earth magnetism noise jamming, avoids driving misoperation.It is worthy of note that the self-shield of IPM mainly is at the non-repeatability transient fault; therefore must be when fault output be arranged; in time cut off circuit; be to realize (as Fig. 5 (b)) by following measure in this device: the 7th pin that the fault output signal of IPM is received TMS320LF2407 by optocoupler extends out 26 pins into P1; DSP in time is set to high-impedance state with all task manager output pins when guaranteeing that IPM breaks down, thereby avoids serious accident.

Fig. 6 is the current sampling circuit schematic diagram, it is input as the output signal of two-phase Hall current sensor, Hall current sensor is powered by positive and negative 15 DC voltage-stabilizing electric current, its input is that any two phase conductors with the U of motor, V, W pass around two circles from the center of Hall current sensor, corresponding output will induce the current signal of 1000:1, be defined as A mutually with B mutually, as the input of current sampling circuit, A, B phase current sampling circuit are identical.Adjustable resistance 503 is used for the conditioning signal amplitude among the figure, adjustable resistance 202 is to be used for the side-play amount of conditioning signal, by the adjusting to these two resistance, signal can be adjusted between 0～3.3V, again it is sent into the AD conversion pin of DSP, select 23 and 24 pins of P2 on the EVM at this.Wherein voltage-stabiliser tube is for the signal that prevents to send into DSP surpasses 3.3V, and damages DSP.Amplifier adopts OP07, connects the voltage of positive and negative 15V, at the indirect decoupling capacitor of voltage and ground.Circuit input end connects capacitance-resistance filter, makes current sampling signal more accurate.

The voltage sampling circuit schematic diagram as shown in Figure 7, its voltage output signal connects the AD conversion pin of DSP, selects 5 pins of P2 on the EVM at this.The Hall voltage transducer is connected on rectifier output end, detect DC bus-bar voltage, its principle is that original edge voltage is converted to primary current by former limit resistance, and magnetic flux that this electric current produces and Hall voltage balance each other by the magnetic flux that secondary coil produced through amplifying the secondary current that produces.Secondary current accurately reflects original edge voltage.All conversion of signals of sampling need be arrived in the DSP acceptable range of voltage values, the A/D convertor circuit that is had by DSP inside samples participation computing after the processing in the DSP.

This device is controlled by the control program that embeds dsp processor, and its control procedure is carried out according to the following steps:

Step 1, initialization;

Step 2, rotor initial alignment;

Step 3, permission INT1, INT2 interrupts;

Step 4, startup T1 underflow are interrupted;

Step 5, interrupt latency;

Step 6, T1 Interrupt Process;

Step 7, protection Interrupt Process;

Step 8, end.

Wherein protect the Interrupt Process process steps as follows:

Step 1 is forbidden all interruptions;

Step 2 is blocked PIM;

Step 3 interrupts returning.

T1 Interrupt Process process steps is as follows:

Step 1 keeps the scene intact;

Step 2 current-voltage sampling;

Step 3 current and voltage data is handled;

Step 4 magnetic linkage is estimated;

Step 5 torque is estimated;

Whether speed regulation of step 6 is to enter step 7, otherwise enters step 9;

Step 7 velocity estimation;

Step 8 speed PI regulates;

Step 9 magnetic linkage and torque Sliding mode variable structure control device are regulated;

Step 10 space vector modulation;

It is on-the-spot that step 11 is recovered;

Step 12 interrupts returning.

Initialization comprises closes all interruptions, DSP initialization, initialization of variable, task manager initialization, AD initialization.

Realize the control of the no transducer of direct torque control, the initial position of rotor is to need to determine, pass to the direct current of a known dimensions to the stator of motor, make stator produce a constant magnetic field like this, the stationary magnetic field of this magnetic field and rotor interacts, force rotor to forward the position that two magnetic linkages overlap to and stop, thereby obtain the initial phase of rotor.

Just can open corresponding interruption after the rotor initial alignment, enter interrupt latency, wait for the generation of interrupt event, will jump to corresponding interrupt service subroutine and carry out corresponding calculated and processing according to the interrupt vector table that defines after interrupting having produced.

Interrupt service subroutine comprises protection interruption subroutine and T1 underflow interrupt service subroutine.

What protect interrupt response is the guard signal of IPM, belongs to external interrupt, broken height among the priority ratio timer T1.IPM can send guard signal automatically when abnormal conditions such as overcurrent, overvoltage take place, this signal is received the power drive protection pin PDPINTA of DSP through conversion.In case there are abnormal conditions to take place, DSP can enter protection interrupt response subprogram, at first forbids all interruptions, makes motor stall at once blocking PWM output then, plays the effect of protection motor and IPM.

Timer adopts the increase and decrease counting mode, subtract to produce when counting up to 0 and interrupt when timer T1 finishes the whole cycle, simultaneously timer begin to increase next time/subtract the count cycle.Timer T1 underflow interrupts belonging to interrupt class INT2, opens the corresponding mask bit of interrupt mask register IMR and EVAIMRA when the initialization timing device.When central broken hair was given birth to, CPU pointed to the appropriate address of interrupt vector table, and set interrupt identification IFR and EVAIFRA, have no progeny in the CPU response, jump to the general interrupts service routine of appointment, interrupt identification IFR automatic clear, and set interrupt mode position INTM, forbid other all maskable interruptions.The interrupt flag bit of EVAIFRA needs software to remove.

In T1 underflow interruption subroutine, mainly finish Speedless sensor sliding moding structure direct torque control algolithm.

Enter and at first keep the scene intact after interrupting, start the AD conversion then, the electric current of being sent back by peripheral circuit, magnitude of voltage are collected in the middle of the DSP.Gathering the data of returning at first is in the result register (RESUTLx) that is stored in separately, from result register read current, magnitude of voltage, electric current is carried out coordinate transform obtain biphase current value under the rest frame.Enter for the first time when interrupting, because rotor has carried out initial alignment, so can obtain the value of two phase voltages under the rest frame first time.The value that has obtained current/voltage just can be carried out the estimation of magnetic linkage, and then carries out torque and estimate.The requirement of response speed is far from magnetic linkage to speed ring and the torque ring is high like that to the requirement of response speed, stipulates that therefore the T1 underflow interrupts all will carrying out the adjusting of magnetic linkage and torque each time, and interrupt just carrying out the estimation and the PI adjusting of a speed per 20 times.Therefore before admission velocity estimation and PI adjusting, to judge whether to want admission velocity to regulate once.Velocity estimation adopts following algorithm to realize:

${\mathrm{\&omega;}}_r(k+1)={\mathrm{\&omega;}}_s(k+1)=\frac{(V_{\mathrm{\&beta;}}\left(k\right)-i_{\mathrm{\&beta;}}\left(k\right)){\mathrm{\&psi;}}_{\mathrm{\&alpha;}}\left(k\right)-(V_{\mathrm{\&alpha;}}\left(k\right)-i_{\mathrm{\&alpha;}}\left(k\right)){\mathrm{\&psi;}}_{\mathrm{\&beta;}}\left(k\right))}{{\mathrm{\&psi;}}_{\mathrm{\&alpha;}}{\left(k\right)}^2+{\mathrm{\&psi;}}_{\mathrm{\&beta;}}{\left(k\right)}^2}$

Subscript α in the formula, β represent static two coordinate systems;

ω _r(k+1)---the k+1 time spinner velocity estimated value;

ω _s(k+1)---the k+1 time stator field rotary speed estimated value;

V---voltage; I---electric current; ψ---stator magnetic linkage;

K---the k time sampling or result of calculation.Obtained the velocity estimation value, obtained velocity error doing difference, carried out speed PI and regulate, the set-point of output torque with the velocity setting value.

After the velocity estimation value obtains the position of rotor is estimated can realize with following formula:

θ _r(k+1)＝θ _r(k)+ω _r(k)ΔT

By the DC bus-bar voltage that AD samples and, looking into rotor-position estimation θ _rThe sine and cosine table, obtain (α, β) V under the coordinate system _α(k+1) and V _β(k+1) value is to be used for the velocity estimation of rotor next time.

Calculate the set-point and the estimated value of magnetic linkage torque according to the current value of given galvanic current value and current sampling circuit sampling, use the Sliding mode variable structure control device that these two amounts are regulated.

The input of Sliding mode variable structure control device is respectively the error of magnetic linkage and torque, is output as α under the rest frame, the component of voltage of β axle.

${\underset{\&OverBar;}{V}}_{\mathrm{ref}}=\left[\begin{array}{c}{V}_{\mathrm{\&alpha;}}(k+1)\\ V_{\mathrm{\&beta;}}(k+1)\end{array}\right]=-{\underset{\&OverBar;}{D}}^{-1}\left(k\right)\left[\begin{array}{cc}{\mathrm{\&mu;}}_1& 0\\ 0& {\mathrm{\&mu;}}_2\end{array}\right]\left[\begin{array}{c}\mathrm{sign}\left(S_1\right)\\ \mathrm{sign}\left(S_2\right)\end{array}\right]$

Wherein

${\underset{\&OverBar;}{D}}^{-1}\left(k\right)=\frac{1}{\left|D\right|}\left[\begin{array}{cc}-{2\mathrm{\&psi;}}_{\mathrm{\&beta;}}\left(k\right)& \frac{3}{2}p(\frac{{\mathrm{\&psi;}}_{\mathrm{\&alpha;}}\left(k\right)}{L_s}-i_{\mathrm{\&alpha;}}\left(k\right))\\ {2\mathrm{\&psi;}}_{\mathrm{\&alpha;}}\left(k\right)& -\frac{3}{2}p(i_{\mathrm{\&beta;}}\left(k\right)-\frac{{\mathrm{\&psi;}}_{\mathrm{\&beta;}}\left(k\right)}{L_s})\end{array}\right]$

$\left|D\right|=3p[({\mathrm{\&psi;}}_{\mathrm{\&alpha;}}\left(k\right)i_{\mathrm{\&alpha;}}\left(k\right)+{\mathrm{\&psi;}}_{\mathrm{\&beta;}}{\left(k\right)i}_{\mathrm{\&beta;}}\left(k\right))-\frac{1}{L_s}(({\mathrm{\&psi;}}_{\mathrm{\&alpha;}}{\left(k\right)}^2+{\mathrm{\&psi;}}_{\mathrm{\&beta;}}{\left(k\right)}^2)]$

$\mathrm{sign}\left(S_i\right)=\left\{\begin{array}{c}1,S_i>{\mathrm{\&lambda;}}_i\\ -1,S_i<{-\mathrm{\&lambda;}}_i\\ \frac{S_i}{{\mathrm{\&lambda;}}_i},\left|S_i\right|<{\mathrm{\&lambda;}}_i\end{array}\right.$

$\left\{\begin{array}{c}{\mathrm{<figure-callout\; id="1"\; label="S"\; filenames="A20061004584900161.png,A20061004584900171.png"\; state="\{\{state\}\}">S</figure-callout>}}_1=e_T\left(k\right)+{\mathrm{<figure-callout\; id="1"\; label="K"\; filenames="A20061004584900161.png,A20061004584900171.png"\; state="\{\{state\}\}">K</figure-callout>}}_1\underset{i=0}{\overset{k}{\mathrm{\&Sigma;}}}{e}_T\left(i\right)-e_T\left(0\right)\\ {\mathrm{<figure-callout\; id="2"\; label="S"\; filenames="A20061004584900161.png,A20061004584900171.png"\; state="\{\{state\}\}">S</figure-callout>}}_2=e_{\mathrm{\&psi;}}\left(k\right)+{\mathrm{<figure-callout\; id="2"\; label="K"\; filenames="A20061004584900161.png,A20061004584900171.png"\; state="\{\{state\}\}">K</figure-callout>}}_2\underset{i=0}{\overset{k}{\mathrm{\&Sigma;}}}{e}_{\mathrm{\&psi;}}\left(i\right)-e_{\mathrm{\&psi;}}\left(0\right)\end{array}\right.$

P in the formula---motor number of pole-pairs; L _s---the stator winding inductance; μ ₁, μ ₂---the sliding formwork coefficient;

I=1,2, λ _i---the width of sliding formwork BL (Boudary layer);

K ₁, K ₂---the sliding formwork gain.

Voltage given value under the rest frame that obtains is the input of space vector modulation (SVPWM), utilize space vector modulation (SVPWM) technology to generate the action time of adjacent two voltage vectors again, change into corresponding value and write three comparand registers (CMPRx), export corresponding pwm signal and drive IPM, finally carry out the SERVO CONTROL of motor.

Claims (7)

1. a sliding moding structure direct torque servo-driver comprises commutation inversion output circuit, control circuit and three parts of controlling object; The commutation inversion output circuit comprises rectification filtering unit and IPM inversion unit; Control circuit comprises dsp processor, IPM isolation drive and protective circuit, current sampling circuit, voltage sampling circuit; Controlling object is a three-phase permanent-magnetic synchronous motors; The input of IPM inversion unit main power source P, N end links to each other with rectification circuit output; voltage sampling circuit is gathered voltage difference between P, N; IPM lead-out terminal U, V, W link to each other with permagnetic synchronous motor; V, W link to each other with the two-way current sampling circuit by two Hall current sensors again; 16 road control terminals of IPM link to each other with IPM isolation drive protective circuit, and the input of IPM isolation drive and the output of protective circuit and electric current, voltage sampling circuit output link to each other with dsp processor.

2. a kind of sliding moding structure direct torque servo-driver according to claim 1, it is characterized in that isolating transformer is connected with voltage regulator after two contact K are connected to rectifier bridge in the described commutation inversion output circuit, the anode of diode is connected to the N end of IPM main power source in the rectifier bridge, the negative electrode of diode is connected to the P end of IPM main power source, the three-phase current of IPM output is by lead-out terminal U, V, W is connected to threephase motor.

3. a kind of sliding moding structure direct torque servo-driver according to claim 1; it is characterized in that dsp processor adopts TMS320LF2407 assessment version in the described control circuit; the address bus of TMS320LF2407 assessment version meets static memory U3 respectively; enlarging P3 outside the address bus of U4 and the address; the data/address bus of TMS320LF2407 assessment version meets U3 respectively; enlarging P3 outside the data/address bus of U4 and the address; the read-write enable pin of TMS320LF2407 assessment version meets U3 respectively; 17 of U4; 41 pins; the program space gating pin of TMS320LF2407 assessment version connects 6 pins of U3; the data space gating pin of TMS320LF2407 assessment version connects 6 pins of U4; the JTAG pin of TMS320LF2407 assessment version meets P5; P5 links to each other with an end of simulator; the other end links to each other with PC by LPT; the analog-to-digital conversion pin of TMS320LF2407 assessment version meets 23 of the outer enlarging P2 of simulation respectively; 24; 5; 6; 7; 8; 9; 10; 11; 12; 13; 14; 15; 16; 19; 20 pins; the PWM interface of TMS320LF2407 assessment version task manager diffuses into 3 of P1 outward; 4; 5; 6; 7; 8; 12; 13; 14; 9; 10; 11; 21; 22; 24 pins; 20 of P4; 25; 26; 27; 29 pins; 1 pin of the external 15M crystal oscillator of the 123 pins U22 of TMS320LF2407 assessment version; the analog references power pin 116 of TMS320LF2407 assessment version; 117 connect 4 and 11 pins of U19 respectively; the digital reference power pin 29 of TMS320LF2407 assessment version; 50; 86; 129; 4; 42; 67; 77; 95; 141 meet 17 of 3.3V voltage source module U12; 18 and 19 pins; 28 of TMS320LF2407 assessment version; 49; 85; 128; 3; 41; 66; 76; 94; 125; it digitally is 9 and 10 pins of U12 that 140 pins connect; outer 3 of the interface P1 that draws together of TMS320LF2407 assessment version; 4; 5; 6; 7; 8 pins directly receive IPM isolation drive and protective circuit input PORT6 1; 3; 4; 5; 6; 7; 7 pins that the fault output signal of IPM is received TMS320LF2407 assessment version by optocoupler extend out 26 pins into P1; TMS320LF2407 assessment version extends out 23 of interface P2 and is connected A respectively with 24 pins; the output of B biphase current sample circuit, TMS320LF2407 assessment version extend out the output of the 5 pins connection voltage sampling circuit of interface P2.

4. the described a kind of sliding moding structure direct torque servo-driver of claim 1 is characterized in that the control procedure of this device is carried out according to the following steps:

Step 1, initialization;

Step 2, rotor initial alignment;

Step 3, permission INT1, INT2 interrupts;

Step 4, startup T1 underflow are interrupted:

Step 5, interrupt latency;

Step 6, T1 Interrupt Process;

Step 7, protection Interrupt Process;

Step 8, end.

5. a kind of sliding moding structure direct torque servo-driver according to claim 4 is characterized in that protection Interrupt Process process is carried out according to the following steps in the step 7:

Step 1, forbid all interruptions;

Step 2, blockade PIM;

Step 3, interruption are returned.

6. a kind of sliding moding structure direct torque servo-driver according to claim 4 is characterized in that T1 Interrupt Process process is carried out according to the following steps in the step 6:

Step 1, keep the scene intact;

Step 2, current-voltage sampling;

Step 3, current and voltage data are handled;

Step 4, magnetic linkage are estimated;

Step 5, torque are estimated;

Step 6, whether speed regulation are to enter step 7, otherwise enter step 9;

Step 7, velocity estimation;

Step 8, speed PI regulate;

Step 9, magnetic linkage and torque Sliding mode variable structure control device are regulated;

Step 10, space vector modulation;

Step 11, recovery scene;

Step 12, interruption are returned.

7. a kind of sliding moding structure direct torque servo-driver according to claim 6, it is characterized in that magnetic linkage and the adjusting of torque Sliding mode variable structure control device in the step 9, the input of Sliding mode variable structure control device is respectively the error of magnetic linkage and torque, be output as α under the rest frame, the component of voltage of β axle, calculate according to following formula:

${\underset{\&OverBar;}{V}}_{\mathrm{ref}}=\left[\begin{array}{c}{V}_{\mathrm{\&alpha;}}(k+1)\\ V_{\mathrm{\&beta;}}(k+1)\end{array}\right]=-{\underset{\&OverBar;}{D}}^{-1}\left(k\right)\left[\begin{array}{cc}{\mathrm{\&mu;}}_1& 0\\ 0& {\mathrm{\&mu;}}_2\end{array}\right]\left[\begin{array}{c}\mathrm{sign}\left(S_1\right)\\ \mathrm{sign}\left(S_2\right)\end{array}\right]$

Wherein

${\underset{\&OverBar;}{D}}^{-1}\left(k\right)=\frac{1}{\left|D\right|}\left[\begin{array}{cc}-2{\mathrm{\&psi;}}_{\mathrm{\&beta;}}\left(k\right)& \frac{3}{2}p(\frac{{\mathrm{\&psi;}}_{\mathrm{\&alpha;}}\left(k\right)}{L_s}-i_{\mathrm{\&alpha;}}\left(k\right))\\ 2{\mathrm{\&psi;}}_{\mathrm{\&alpha;}}\left(k\right)& -\frac{3}{2}p(i_{\mathrm{\&beta;}}\left(k\right)-\frac{{\mathrm{\&psi;}}_{\mathrm{\&beta;}}\left(k\right)}{L_s})\end{array}\right]$

$\left|D\right|=3p[({\mathrm{\&psi;}}_{\mathrm{\&alpha;}}\left(k\right)i_{\mathrm{\&alpha;}}\left(k\right)+{\mathrm{\&psi;}}_{\mathrm{\&beta;}}\left(k\right)i_{\mathrm{\&beta;}}\left(k\right))-\frac{1}{L_s}\left(({\mathrm{\&psi;}}_{\mathrm{\&alpha;}}{\left(k\right)}^2+{\mathrm{\&psi;}}_{\mathrm{\&beta;}}{\left(k\right)}^2)\right)]$

$\mathrm{sign}\left(S_i\right)=\left\{\begin{array}{c}1,S_i>{\mathrm{\&lambda;}}_i\\ -1,S_i<-{\mathrm{\&lambda;}}_i\\ \frac{S_i}{{\mathrm{\&lambda;}}_i},\left|S_i\right|<{\mathrm{\&lambda;}}_i\end{array}\right.$

$\left\{\begin{array}{c}{S}_1=e_T\left(k\right)+K_1\underset{i=0}{\overset{k}{\mathrm{\&Sigma;}}}{e}_T\left(i\right)-e_T\left(0\right)\\ S_2=e_{\mathrm{\&psi;}}\left(k\right)+K_2\underset{i=0}{\overset{k}{\mathrm{\&Sigma;}}}{e}_{\mathrm{\&psi;}}\left(i\right)-e_{\mathrm{\&psi;}}\left(0\right)\end{array}\right.$

P in the formula---motor number of pole-pairs; L _s---the stator winding inductance; μ ₁, μ ₂---the sliding formwork coefficient;

ψ---stator magnetic linkage; K---the k time sampling or result of calculation;

I=1,2, λ _i---the width of sliding formwork BL (Boudary layer);

K ₁, K ₂---the sliding formwork gain.

Images