JP2022095725A - Inverter drive method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a three-phase PWM inverter drive method that can reduce heat generation and power loss.

SOLUTION: Three H-bridges of a dual three-phase inverter each consist of a fixed-potential leg and a PWM leg that alternate every 180 degrees of electrical angle, and the fixed-potential leg has an upper arm switch that is always on. Current supply period of the three H-bridges do not overlap each other under an operating condition in which a power supply current is less than a predetermined value. The two relatively short current supply periods of the three H-bridges preferentially overlap under an operating condition in which the power supply current exceeds the predetermined value.

SELECTED DRAWING: Figure 1

COPYRIGHT: (C)2022,JPO&INPIT

Description

The present invention relates to an inverter drive method, and more particularly to a pulse width modulation method for a three-phase inverter for driving a variable speed motor.

In high power applications such as variable speed motor applications and power conditioner applications, it is common to employ a pulse width modulation (PWM) three-phase inverter capable of supplying a three-phase sinusoidal current. A three-phase PWM inverter consisting of three legs is called a single inverter. A three-phase PWM inverter with three H-bridges is called a dual inverter. In motor applications, a single inverter is typically connected to a star-connected three-phase coil, and the two three-phase inverters of the dual inverter are connected to an open-ended three-phase coil (double-ended three-phase coil). Each leg of the voltage type inverter consists of an upper arm switch and a lower arm switch connected in series. Each arm switch has a transistor and an antiparallel diode.

Each leg of the three-phase inverter controlled by the pulse width modulation method is controlled for each PWM cycle period corresponding to one cycle of the carrier signal. Each PWM cycle period has a current supply period in which the pulsed power supply current flows between the DC power supply and the electric load. The current supply period is called the main voltage vector. The PWM cycle period other than the current supply period is called the freewheeling period. The freewheeling period is called the zero vector. The rotationally combined voltage vector corresponding to the substantially sinusoidal waveform voltage is formed by using six main voltage vectors and two zero vectors. This combined voltage vector corresponds to the vector sum of two main voltage vectors adjacent to each other. Therefore, each PWM cycle period includes two types of current supply periods.

Patent Document 1 describes a square wave pulse width modulation that forms a combined voltage vector using six principal voltage vectors and six subvoltage vectors. According to this square wave pulse width modulation, each PWM cycle period contains only one of one principal voltage vector and one subvoltage vector. In other words, each PWM cycle period includes only one type of current supply period. The secondary voltage vector applies the power supply voltage only to the two phase coils of the star three-phase coil. Therefore, the power supply current is not supplied to the remaining one leg.

Patent Document 2 describes a technique for switching the number of turns of a motor using a dual inverter including a short circuit. The dual inverter consists of two 3-phase inverters, each of which is PWM controlled. The three AC output terminals of one three-phase inverter are short-circuited by a short-circuit circuit. As a result, the number of turns of the stator coil is switched.

Patent Document 3 describes a dual three-phase inverter connected to a double-ended three-phase coil. The dual three-phase inverter consists of three H-bridges. This dual three-phase inverter consisting of two three-phase inverters operating in opposite directions has a pulse width modulation mode and a pulse drive mode. In pulse width modulation mode, the two three-phase inverters output opposite pulse width modulated three-phase voltages in PWM mode. In the pulse drive mode, the two three-phase inverters output square wave voltages with opposite waveforms to each other. In other words, the two legs of the H-bridge each become a PWM leg in the pulse width modulation mode and each become a fixed potential leg in the pulse drive mode.

Patent Document 4 proposes first and second single three-phase inverters that drive two motors separately. The first inverter is PWM controlled by the first carrier signal, and the second inverter is PWM controlled by the second carrier signal. The first carrier signal is shifted half a cycle from the second carrier signal. As a result, there may be a case where the pulsed power supply current supplied by the DC power supply to the first inverter does not overlap with the pulsed power supply current supplied by the DC power supply to the second inverter.

The problem of Patent Document 1 will be explained. Patent Document 1 forms a combined voltage vector using only a main voltage vector or a sub voltage vector. Therefore, the vibration, noise, and iron loss of the motor increase. In other words, in Patent Document 1, since the main voltage vector equivalent current supply period and the sub voltage vector equivalent current supply period are not arranged within one PWM cycle period, the combined voltage vector cannot rotate smoothly.

The problem of Patent Document 2 will be explained. Each of the three H-bridges PWM controls different independent single-phase electrical loads. Each H-bridge consists of a first PWM leg and a second PWM leg, respectively. The H-bridge has four output states (00, 01, 10, and 11). The output states (00) and (11) are freewheeling periods corresponding to zero vectors. The output state (01) corresponds to the current supply period for supplying the power supply current from the first PWM leg to the second PWM leg. The output state (10) is a current supply period for supplying a power supply current from the second PWM leg to the first PWM leg.

Therefore, each PWM cycle period includes both two current supply periods (01, 10) for supplying power supply currents in opposite directions. The substantial current supply period is the difference between the first current supply period and the second current supply period, and the duty ratio is equal to the difference / PWM cycle period. Eventually, two current supply periods with opposite voltage application directions are formed within each PWM cycle period, resulting in increased current and voltage ripple. This problem is called the reverse voltage application problem.

The problem of Patent Document 3 will be explained. Patent Document 3 does not describe at all that an H-bridge is formed by one PWM leg and one fixed potential leg, and the fixed potential leg is switched every 180 degrees of electric angle.

The problem of Patent Document 4 will be explained. It is impossible for Patent Document 4 to reduce the amplitude of the pulsed power supply current supplied by the DC power supply to one three-phase motor.

Japanese Patent No. 4053368 Japanese Patent No. 4906836 Japanese Unexamined Patent Publication No. 2006-149145 Japanese Patent No. 6184712

One object of the present invention is to provide a three-phase PWM inverter drive method capable of reducing heat generation and power loss.

According to the present invention, the three H-bridges of a dual three-phase PWM inverter consist of one PWM leg and one fixed potential leg, respectively. Either one of the upper arm switch and the lower arm switch of the fixed potential leg is always turned on. The fixed potential leg is switched every 180 degrees of electrical angle. In other words, half the legs of each H-bridge are not PWM switched. As a result, this dual three-phase inverter can realize half the switching loss as compared with the conventional dual three-phase inverter in which the H bridge consists of two PWM legs. Furthermore, this dual inverter does not cause the reverse voltage application problem described above.

The dual three-phase inverter, which requires two three-phase inverters, is worried about an increase in manufacturing cost. However, the dual three-phase inverter can have almost twice the voltage utilization rate as the single three-phase inverter. Therefore, this dual three-phase inverter can have a total semiconductor chip area that is totally equal to that of a conventional single three-phase inverter.

According to the first aspect, the current supply periods of the three H-bridges are arranged within a common PWM cycle period so as not to overlap as much as possible. This makes it possible to reduce the power loss of the DC power supply. This aspect can also be applied to a conventional dual three-phase inverter.

According to the second aspect, the two relatively short current supply periods of the three H-bridges preferentially overlap. The length of the current supply period is approximately proportional to the current amplitude. Therefore, it is possible to suppress an increase in power loss of the upper arm switch of the fixed potential leg and the DC power supply due to the overlap of the current supply periods. This aspect can also be applied to a conventional dual three-phase inverter.

According to the third aspect, the current supply period of one H-bridge that supplies the generated current to the DC power supply exceeds the current supply period of the other H-bridge that supplies the motor drive current to the inductive single-phase electric load. Wrapped. As a result, the amplitude value of the power supply current supplied by the DC power supply to the inverter can be reduced. As a result, the power loss of the DC power supply can be reduced. This aspect can also be applied to a conventional dual three-phase inverter.

According to the fourth aspect, each of the pair of power supply lines connecting the DC power supply and the inverter is individually covered with a metal tube. Further, each end of the two metal tubes is connected by a first crossover and the other ends of the two metal tubes are connected by a second crossover. As a result, the surge voltage of the power supply line can be reduced. This aspect can also be applied to a conventional inverter.

According to the fifth aspect, each of the three or more motor cables connecting the inverter and the stator coil of the motor is individually covered with a metal tube. Further, one end of each metal tube is connected by a first crossover, and the other end of each metal tube is connected by a second crossover. This makes it possible to reduce the surge voltage of the motor-cable. This aspect can also be applied to a conventional inverter.

According to a sixth aspect, the fixed potential leg has an upper arm switch that is always on. The power supply current is controlled by switching the lower arm switch of the PWM leg. Therefore, when the power supply current decreases due to the OFF of the lower arm switch of the PWM leg, the upper arm switch of the fixed potential leg is ON. As a result, the voltage induced in the + power line connected to the upper arm switch is reduced. The three-phase inverter of the present invention in which the upper arm switch of the fixed potential leg is always on is called an upper arm conduction type three-phase inverter.

According to the seventh aspect, the off operation of the upper arm switch of the fixed potential leg performed for switching the fixed potential leg is performed immediately after the freewheeling period. In other words, the upper arm switch of the fixed potential leg cuts off the freewheeling current instead of the power supply current for switching the fixed potential leg. Freewheeling current flows through the + busbar connecting the two upper arm switches of the H-bridge. This + bus bar, which is much shorter than the + power line through which the power supply current flows, has a low inductance value. After all, the voltage generated by turning off the upper arm switch for switching the fixed potential leg can be significantly reduced.

The voltage is further explained. + The reduction of the voltage, which is the surge voltage of the power supply line, realizes the reduction of the withstand voltage of the semiconductor element of the inverter and the smoothing capacitor. As a result, manufacturing cost and resistance loss can be reduced. + Power line voltage is classified into external surge voltage and internal surge voltage. The external surge voltage is generated by a sharp decrease in the power supply current flowing through the inductance of the + power supply line connecting the DC power supply and the inverter. The internal surge voltage is generated by a sharp decrease in the freewheeling current flowing through the + busbar that connects each upper arm switch of the inverter. After all, by executing the off operation of the upper arm switch of the fixed potential leg during the freewheeling period, a low internal surge voltage can be generated instead of the high external surge voltage.

FIG. 1 is a schematic wiring diagram showing a dual inverter of an embodiment. FIG. 2 is a timing chart showing an output voltage waveform and an output current waveform of one H-bridge. FIG. 3 is a timing chart showing the state of one H-bridge in the PWM cycle period of the half-wave period. FIG. 4 is a timing chart showing the state of one H-bridge in the PWM cycle period of the negative half-wave period. FIG. 5 is a schematic wiring diagram showing the state of one H-bridge in the current supply period of the first phase period. FIG. 6 is a schematic wiring diagram showing the state of one H-bridge in the freewheeling period of the first phase period. FIG. 7 is a schematic wiring diagram showing the state of one H-bridge in the current supply period of the second phase period. FIG. 8 is a schematic wiring diagram showing the state of one H-bridge in the freewheeling period of the second phase period. FIG. 9 is a schematic wiring diagram showing the state of one H-bridge in the current supply period of the third phase period. FIG. 10 is a schematic wiring diagram showing the state of one H-bridge in the freewheeling period of the third phase period. FIG. 11 is a schematic wiring diagram showing the state of one H-bridge in the current supply period of the fourth phase period. FIG. 12 is a schematic wiring diagram showing the state of one H-bridge in the freewheeling period of the fourth phase period. FIG. 13A is a timing chart showing the operation of the six legs at an electrical angle of 360 degrees. FIG. 13B is a waveform diagram showing fundamental wave components of the output voltages of the two legs of the H-bridge at an electrical angle of 360 degrees. FIG. 14 is a schematic wiring diagram showing a wiring example of the dual inverter. FIG. 15 is a front view of the dual inverter shown in FIG. FIG. 16 is a plan view of the dual inverter shown in FIG. FIG. 17 is a side view of the dual inverter shown in FIG. FIG. 18 is a timing chart for explaining the current distribution method of the dual inverter. FIG. 19 is a schematic wiring diagram showing a motor device that employs a current distribution method. FIG. 20A is a schematic wiring diagram for explaining the surge voltage generated by turning off the upper arm switch of the H bridge. FIG. 20B is a schematic wiring diagram for explaining the surge voltage generated by turning off the lower arm switch of the H bridge.

FIG. 1 is a wiring diagram showing a dual inverter for driving an EV traction motor. This dual inverter has an inverter 1 and an inverter 2 each consisting of a voltage type three-phase inverter. Further, this dual inverter has a short circuit 9. The stator coil of the EV traction motor includes a double-ended three-phase coil including a U-phase coil 3U, a V-phase coil 3V, and a W-phase coil 3W.

The inverter 1 includes a U-phase leg 1U, a V-phase leg 1V, and a W-phase leg 1W. The leg 1U includes an upper arm switch 11 and a lower arm switch 12 connected in series. The leg 1V includes an upper arm switch 13 and a lower arm switch 14 connected in series. The leg 1W includes an upper arm switch 15 and a lower arm switch 16 connected in series. The inverter 2 includes a U-phase leg 2U, a V-phase leg 2V, and a W-phase leg 2W. The three upper arm switches connect one end of each of the phase coils 3U, 3V, and 3W to the + busbar 810. The three lower arm switches connect one end of each of the phase coils 3U, 3V, and 3W to the-busbar 820.

The leg 2U includes an upper arm switch 21 and a lower arm switch 22 connected in series. The leg 2V includes an upper arm switch 23 and a lower arm switch 24 connected in series. The leg 2W includes an upper arm switch 25 and a lower arm switch 26 connected in series. The three upper arm switches connect the other ends of the phase coils 3U, 3V, and 3W to the + busbar 810. The three lower arm switches connect the other ends of the phase coils 3U, 3V, and 3W to the-busbar 820. Each arm switch consists of an antiparallel diode and a transistor.

The leg 1U applies a U-phase voltage VU1 to the phase coil 3U. The leg 2U applies a U-phase voltage VU2 to the phase coil 3U. As a result, the U-phase voltage VU (= VU1-VU2) is applied to the phase coil 3U. The leg 1V applies a V-phase voltage VV1 to the phase coil 3V. The leg 2V applies a U-phase voltage VV2 to the phase coil 3V. As a result, the V-phase voltage VV (= VV1-VV2) is applied to the phase coil 3V. The leg 1W applies a W-phase voltage VW1 to the phase coil 3W. The leg 2W applies a W-phase voltage VW2 to the phase coil 3W. As a result, the W phase voltage VW (= VW1-VW2) is applied to the phase coil 3W. The phase difference between the three phase voltages (VU, VV, and VW) is an electrical angle of 120 degrees.

The short circuit 9 includes a three-phase diode bridge and a short circuit switch. The rectifier composed of the three-phase diode bridge rectifies the three-phase voltage (VU2, VV2, VW2) of the three-phase inverter 2. The pair of DC terminals of the three-phase diode bridge can be short-circuited by a short-circuit switch. The controller 10 PWM controls the three-phase inverters 1 and 2. Further, the controller 10 opens and closes the short-circuit switch. When the short circuit switch is turned off, the three phase coils 3U, 3V and 3W are individually controlled by the three H bridges. This control is called open delta mode. When the short circuit switch is turned on, the three phase coils 3U, 3V and 3W become star-shaped connected three-phase coils. The three-phase inverter 1 controls this star-shaped connection three-phase coil. The three-phase inverter 2 is suspended. This control is called star mode.

Open delta mode is described. In open delta mode, the three H-bridges perform essentially the same control operation. The phase difference between the control operations of each H-bridge is an electrical angle of 120 degrees. In the following, PWM control of a U-phase H bridge consisting of legs 1U and 2U will be described. This control is called upper arm conduction one-sided PWM.

FIG. 2 is a waveform diagram showing each fundamental wave component of the U-phase voltage VU and the U-phase current IU applied to the phase coil 3U from the U-phase H bridge. The U-phase current IU lags the U-phase voltage VU by a predetermined phase angle. One cycle of the phase voltage VU (= electric angle 360 degrees) consists of a positive half-wave period PA and a negative half-wave period PB. The period PA is divided into phase periods P1 and P2. The period PB is divided into phase periods P3 and P4. The legs 1U and 2U are pulse width modulation (PWM) controlled. As the pulse width modulation, space vector pulse width modulation in which the current supply period can be freely arranged is suitable.

The controller 10 sets a current supply period TX having an appropriate length for each PWM cycle period TC which is a control cycle. Therefore, the PWM duty ratio is TX / TC. The PWM cycle period TC excluding the current supply period TX is the freewheeling period TF. In the current supply period TX, the DC power supply supplies the power supply current IP to the stator coil. During the freewheeling period Tf, the freewheeling current IF circulates between the inverter and the stator coil.

3 and 4 show the states of legs 1U and 2U in one PWM cycle period TC, respectively. FIG. 3 shows an example of the state of the legs 1U and 2U in the period PA, and FIG. 4 shows an example of the state of the legs 1U and 2U in the period PB. The period PA is divided into phase periods P1 and P2, and the period PB is divided into phase periods P3 and P4. The phase periods P1 and P3 are periods for regenerating the power supply current from the phase coil 3U to the DC power supply, and the phase periods P2 and P4 are periods for the DC power supply to supply the power supply current IP to the phase coil 3U.

According to the upper arm conduction type one-sided PWM of this embodiment, the H bridge consisting of legs 1U and 2U consists of a fixed potential leg and a PWM leg, and the upper arm switch of the fixed potential leg is always turned on. In the period PA, leg 1U becomes a fixed potential leg and leg 2U becomes a PWM leg. In period PB, leg 1U becomes a PWM leg and leg 2U becomes a fixed potential leg. The fixed potential leg and the PWM leg are alternated every 180 degrees of electrical angle.

FIG. 3 is a timing chart showing one PWM cycle period TC in the period PA. The fixed potential leg 1U to which the upper arm switch 11 is turned on outputs a high level (1). In the current supply period TX, the lower arm switch 22 is turned on and the PWM leg 2U outputs a low level (0). During the freewheeling period TF, the lower arm switch 22 is turned off and the PWM leg 2U outputs a high level (1).

FIG. 4 is a timing chart showing one PWM cycle period TC in the period PB. The fixed potential leg 2U to which the upper arm switch 21 is turned on outputs a high level (1). In the current supply period TX, the lower arm switch 12 is turned on, and the PWM leg 1U outputs a low level (0). During the freewheeling period TF, the lower arm switch 12 is turned off and the PWM leg 1U outputs a high level (1).

5 to 12 are schematic wiring diagrams showing the states of legs 1U and 2U in the phase periods P1-P4. The phase coil 3U has a counter electromotive force Va. FIG. 5 shows the current supply period TX of the phase period P1, and FIG. 6 shows the freewheeling period TF of the phase period P1. A regenerative current IR, which is a negative power supply current IP, flows to the DC power supply 4 during the current supply period TX. The freewheeling current If flowing in the freewheeling period TF stores magnetic energy in the phase coil 3U. FIG. 7 shows the current supply period TX of the phase period P2, and FIG. 8 shows the freewheeling period TF of the phase period P2. The power supply current IP flows to the phase coil 3U during the current supply period TX.

FIG. 9 shows the current supply period TX of the phase period P3, and FIG. 10 shows the freewheeling period TF of the phase period P3. The regenerative current IR flows to the DC power supply 4 during the current supply period TX. The freewheeling current If flowing in the freewheeling period TF stores magnetic energy in the phase coil 3U. FIG. 11 shows the current supply period TX of the phase period P4, and FIG. 12 shows the freewheeling period TF of the phase period P4. The power supply current IP flows to the phase coil 3U during the current supply period TX. FIG. 13A is a timing chart showing the states of the six legs (1U-2W). FIG. 13B is a waveform diagram showing the fundamental wave components of the output voltages (VU1, VU2) output by the two legs 1U and 2U of the U-phase H bridge.

FIG. 14 is a schematic wiring diagram showing a wiring example of the dual three-phase inverter shown in FIG. The short circuit 9 is not shown. The three AC terminals of the three-phase inverter 1 are connected to each end of the phase coil 3U-3W by an AC bus bar 61-63. The three AC terminals of the three-phase inverter 2 are connected to the other ends of the phase coils 3U-3W by AC bus bars 71-73. The + bus bar 810 is connected to the positive electrode terminal of the DC power supply shown in the figure by the + power supply line 81. -The bus bar 820 is connected to the negative electrode terminal of the DC power supply shown in the figure by the power supply line 82.

15-17 show the structure of the dual inverter shown in FIG. FIG. 15 is a front view of the dual inverter. FIG. 16 is a plan view showing a horizontal cross section of the dual inverter, and FIG. 17 is a side view showing a vertical side surface of the dual inverter. -The bus bar 820 has a box shape with two sides omitted. + Busbar 810, busbars 61-63 and 71-73-project horizontally from busbar 820. One opening of the box-shaped busbar 820 is covered with a resin cover 9.

The upper arm switch 11 and the lower arm switch 12 sandwich the bus bar 61. The upper arm switch 13 and the lower arm switch 14 sandwich the bus bar 62. The upper arm switch 15 and the lower arm switch 16 sandwich the bus bar 63. The upper arm switch 21 and the lower arm switch 22 sandwich the bus bar 71. The upper arm switch 23 and the lower arm switch 24 sandwich the bus bar 72. The upper arm switch 25 and the lower arm switch 26 sandwich the bus bar 73. Each lower arm switch is in contact with the busbar 820. Each upper arm switch is in contact with the + bus bar 810. In FIGS. 15-17, each arm switch comprises two semiconductor elements connected in parallel. One of the semiconductor elements is a transistor and the other one is an antiparallel diode.

+ The bus bar 810 is sandwiched by the U-phase upper arm switches 11 and 21. + The bus bar 810 is sandwiched by the V-phase upper arm switches 13 and 23. + The bus bar 810 is sandwiched by the upper arm switches 15 and 25 of the W phase. As a result, the distance of the + bus bar 810 between the upper arm switches 11 and 21 becomes substantially equal to the thickness of the + bus bar 810. The distance of the + busbar 810 between the upper arm switches 13 and 23 is approximately equal to the thickness of the + busbar 810. The distance of the + busbar 810 between the upper arm switches 15 and 25 is approximately equal to the thickness of the + busbar 810.

+ When the freewheeling current If flowing through the bus bar 810 is reduced by turning off the upper arm switch, the line inductance of the + bus bar 810 generates a surge voltage. However, since the length of the + bus bar 810 is extremely short, this surge voltage becomes very low. In addition, the box-shaped-bus bar 820 reduces electromagnetic noise.

As mentioned above, according to this embodiment that employs space vector pulse width modulation (SVPWM), the current supply period TX of each phase can be freely arranged within the PWM cycle period TC. FIG. 18 is a timing chard showing a suitable arrangement example of the three-phase current supply period TX. The controller 10 determines the length of the three current supply periods TX from the calculated phase voltage command value. The controller 10 has six timers for determining the start time point and the end time point of the current supply period TX of each phase. Each timer starts from t0 at the start of the PWM cycle period TC and ends at a predetermined time (t1-t3).

As shown in FIG. 18, the current supply period TX of each phase is arranged in the PWM cycle period TC so as not to overlap each other. This method is called a current distribution method. This current distribution method is adopted under the partial load condition of the motor. When the motor load increases, the current supply periods TX of each phase overlap each other. In this embodiment, two relatively short current supply periods TX are preferentially overlapped. This is because a relatively long current supply period TX means that the amplitude of the phase current is high. This makes it possible to reduce the resistance loss of the DC power supply.

This resistance loss reducing effect will be described with reference to FIG. FIG. 19 is a block diagram showing a traction motor device using a dual three-phase inverter 1X. The DC power supply 4 having the power supply resistance value (r) supplies the power supply current IP to the inverter 1X through the + power supply line 81 and the-power supply line 82. The inverter 1X supplies the U-phase power supply current IUP to the phase coil 3U, supplies the V-phase power supply current IVP to the phase coil 3V, and supplies the W-phase power supply current IWP to the phase coil 3W. Therefore, the power supply current IP is equal to the sum of the three phase power supply currents (IUP, IVP, and IWP). The power supply currents IUP, IVP, and IWP of each phase supplied by the inverter 1X to the stator coil 3 have a pulse current shape. Therefore, the power supply current IP has a pulse shape.

When the power supply currents (IUP, IVP, and IWP) of each phase overlap each other, the resistance loss of the DC power supply 4 becomes (r) (IUP + IVP + IWP) (IUP + IVP + IWP). When the power supply currents (IUP, IVP, and IWP) of each phase do not overlap each other, the resistance loss of the DC power supply 4 is (r) ((IUP) (IUP) + (IVP) (IVP) + (IWP) (IWP). ).

For example, when the V-phase power supply current IVP and the W-phase power supply current IWP each have a relative amplitude value (1), the U-phase power supply current IUP has a relative amplitude value (2). When the power supply currents IUP, IVP, and IWP overlap each other, the resistance loss of the DC power supply 4 becomes (16r). When the power supply currents UP, IVP, and IWP do not overlap each other, the resistance loss of the DC power supply 4 becomes (6r). Therefore, the current distribution method significantly reduces the resistance loss of the DC power supply 4 under the partial load condition.

For example, when the relative amplitude value of the U-phase power supply current IUP is zero, the V-phase power supply current IVP and the W-phase power supply current IWP each have a relative amplitude value (1). When the power supply currents IUP, IVP, and IWP overlap each other, the resistance loss of the DC power supply 4 becomes (4r). When the power supply currents UP, IVP, and IWP do not overlap each other, the resistance loss of the DC power supply 4 becomes (2r). Therefore, the current distribution method significantly reduces the resistance loss of the DC power supply 4 under the partial load condition. After all, the loss of the DC power supply 4 connected to the dual three-phase inverter 1X becomes 37.5% -50% by adopting this current distribution method.

When the sum of the three current supply periods TX is longer than the PWM cycle period TC, the three-phase current supply periods TX shown in FIG. 33 overlap. When this overlap is required, the relatively short two-phase current supply period TX is preferentially overlapped. This is because the short current supply period TX has a lower phase current than the long current supply period TX. This makes it possible to suppress an increase in the resistance loss of the DC power supply 4.

However, the H-bridge driving the inductive electrical load forms a regenerative current during the phase periods P1 and P3 shown in FIG. Therefore, it is preferable that the current supply period of the H bridge of the phase periods P1 and P3 overlaps with the current supply period of the H bridge of the phase periods P2 and P4. This makes it possible to reduce the amplitude of the power supply current IP.

This open delta mode dual inverter is compared to a conventional single three-phase inverter. This dual inverter, which employs a one-sided PWM drive system, has about twice the voltage utilization rate of a conventional single inverter connected to a star-shaped three-phase coil. Therefore, this dual inverter can drive a stator coil having twice the number of turns as compared with a single inverter.

Therefore, the inverters 1 and 2 of the dual inverter can have half the semiconductor chip area as compared with the single inverter. After all, a dual inverter in open delta mode can have a total semiconductor chip area that is approximately equal to that of a conventional single three-phase inverter. As a result, the dual inverter can have substantially the same conduction loss and manufacturing cost as the conventional single inverter.

The effect of the dual inverter of this embodiment will be explained. First, each of the three H-bridges in this dual inverter has a fixed potential leg. This method is called the one-sided PWM method. As a result, this dual inverter can have half the switching loss as compared with the conventional single inverter.

Secondly, this dual inverter can improve the above-mentioned reverse voltage application problem. According to the conventional two-sided PWM method, the H bridge may have a current supply period corresponding to the voltage vector (01) and a current supply period corresponding to the voltage vector (10) in one PWM cycle period TC. According to this one-sided PWM method, two current supply periods corresponding to two voltage vectors in opposite directions are not formed in one PWM cycle period.

Third, in this one-sided PWM method, the upper arm switch of the fixed potential leg is always turned on. As a result, the high voltage generated by the conventional single three-phase inverter can be significantly reduced.

The voltage problem of a conventional double-sided PWM dual inverter or a conventional single inverter will be described with reference to FIG. 20A. The DC power supply 4 is connected to the + bus bar 810 of the U-phase H bridge 1HU through the + power supply line 81, and is connected to the-bus bar 820 of the U-phase H bridge 1HU through the-power supply line 82. The H-bridge 1HU consists of legs 1U and 2U. The phase coil 3U is connected to the leg 1U through the U-phase cable 61 and is connected to the leg 2U through the U-phase cable 71. The + power line 81 has a line inductance of 81L, and the-power line 82 has a line inductance of 82L. + Both ends of the power supply line 81 have parasitic capacitances C1 and C2. -Both ends of the power line 82 have parasitic capacitances C3 and C4. The U-phase cable 61 has a parasitic capacitance C5. The U-phase coil 3U has a parasitic capacitance C6. The U-phase cable 71 has a parasitic capacitance C7.

FIG. 20A shows a state immediately after the upper arm switch 11 through which the power supply current IP flows is turned off. The line inductance 81L generates an electromotive force because the power supply current IP is cut off. This electromotive force supplies a surge current ISU through a series resonant circuit consisting of parasitic capacitors C2 and C1 and a line inductance of 81L. As a result, a high voltage Vr is applied to the upper arm switch 11.

FIG. 20B shows an H-bridge driven by the upper arm conduction type one-sided PWM method of this embodiment. The upper arm switch 11 is always on and the lower arm switch 22 is switched. FIG. 20B shows a state immediately after the lower arm switch 22 through which the power supply current IP flows is turned off. The power supply current IP flowing through the line inductance 81L is reduced because the lower arm switch 22 is turned off, and the potential of the U-phase cable 71 rises rapidly. When the diode of the upper arm switch 21 is turned on by the potential rise of the U-phase cable 71, the freewheeling current If flows into the + bus bar 810. This freewheeling current If circulates through the upper arm switch 11.

The power supply current IP flowing through the line inductance 81L suddenly decreases immediately after the lower arm switch 22 is turned off, and the line inductance 81L generates an electromotive force. This electromotive force supplies the surge current ISU through a series resonant circuit consisting of parasitic capacitors C5-C7 and C1. As a result, the surge current ISU flowing through the parasitic capacitor C2 becomes very low. As a result, the voltage Vr is significantly reduced.

Next, the voltage generated by the + bus bar 810 will be described. When the upper arm switch 21 of the PWM leg 2U shown in FIG. 6 is turned off, the upper arm switch 21 turns off the freewheeling current If flowing from the + bus bar 810. Similarly, when the upper arm switch 11 of the PWM leg 1U shown in FIG. 10 is turned off, the upper arm switch 11 turns off the freewheeling current If flowing from the + bus bar 810. Therefore, the line inductance of the + busbar 810 generates a voltage when the upper arm switch of the PWM leg is turned off. The + bus bar 810 is very short compared to the + power line 81. Therefore, the voltage generated by the + bus bar 810 is very low.

However, dual inverters have the potential to generate voltage when switching between fixed potential legs. This problem will be described with reference to FIG. The current IU is not zero at each start of the period PB during which the fixed potential leg is switched from leg 1U to leg 2U. Similarly, the current IU is not zero at each start of the period PA during which the fixed potential leg is switched from leg 2U to leg 1U. Therefore, the upper arm switch 11 is turned off when the leg 1U is switched from the fixed potential leg to the PWM leg. Similarly, the upper arm switch 21 is turned off when the leg 2U is switched from the fixed potential leg to the PWM leg. As a result, a voltage may be generated.

This problem is solved by performing the off operation of the upper arm switch for switching from the fixed potential leg to the PWM leg during the freewheeling period TF. In other words, the off operation of the upper arm switch of the fixed potential leg is executed in a state where the freewheeling current If circulates through the + bus bar 810 connecting the legs 1U and 2U. As a result, the upper arm switch turns off the freewheeling current If. As mentioned above, the + busbar 810 has a very low line inductance, so the voltage is significantly reduced.

The inverter of the present invention and a driving method thereof disclose a novel novel inverter system including an elevator, an air conditioner, a variable speed motor application such as a traction motor and a marine motor, and a solar cell power conditioner and a controller for controlling the solar cell power conditioner. .. This inverter system is summarized below.

First, the controller controls a dual three-phase inverter consisting of three H-bridges. One PWM leg and one fixed potential leg of each H-bridge are switched every 180 degrees of electrical angle. Thereby, the switching loss can be reduced. Further, it is possible to prevent the reverse voltage application problem in which two current supply periods that generate phase voltages in opposite directions are formed within one PWM cycle period. Preferably, the controller has a fixed potential leg upper arm switch that is constantly turned on. As a result, the surge voltage due to the off of the upper arm switch can be reduced.

Second, the controller is placed within a common PWM cycle period so that the current supply periods of the three phases of the dual three-phase inverter do not overlap as much as possible. This makes it possible to reduce the power loss of the DC power supply.

Third, the dual three-phase inverter preferentially overlaps the two-phase current supply periods with relatively low phase currents. This makes it possible to reduce the power loss of the DC power supply.

Fourth, the dual three-phase inverter preferentially overlaps the current supply period of the phase that outputs the generated current with the current supply period of the other phase. This makes it possible to reduce the power loss of the DC power supply.

Fifth, each of the three H-bridges has a pair of upper arm switches that sandwich the + busbar connected to the positive electrode terminal of the DC power supply. As a result, the surge voltage can be reduced.

Claims (4)

For each PWM cycle period consisting of a current supply period for supplying a power supply current from a DC power supply to the stator coil of a 3-phase variable speed motor and a freewheeling period for supplying a freewheeling current to the stator coil. In the inverter drive method in which the three-phase inverter is controlled by the pulse width modulation (PWM) method and the winding switching of the stator coil is not performed.
The three-phase inverter consists of three H bridges that form the stator coil and separately supply phase currents to the three phase coils that are independent of each other.
Each H-bridge consists of a PWM leg that is switched during the PWM cycle period and a fixed potential leg that outputs a constant voltage during the PWM cycle period.
The PWM leg and the fixed potential leg are alternated every 180 degrees of electrical angle.
The fixed potential leg is an inverter driving method characterized by having an upper arm switch that is always turned on. The inverter driving method according to claim 1, wherein the upper arm switch of the fixed potential leg performs an off operation for the replacement of the PWM leg and the fixed potential leg at or immediately after the freewheeling period. For each PWM cycle period consisting of a current supply period for supplying a power supply current from a DC power supply to the stator coil of a 3-phase variable speed motor and a freewheeling period for supplying a freewheeling current to the stator coil. In the inverter drive method in which the three-phase inverter is controlled by the pulse width modulation (PWM) method and the winding switching of the stator coil is not performed.
The three-phase inverter consists of three H bridges that form the stator coil and separately supply phase currents to the three phase coils that are independent of each other.
An inverter drive method characterized in that two of the relatively short current supply periods of the three H bridges preferentially overlap under operating conditions in which the power supply current exceeds a predetermined value. .. For each PWM cycle period consisting of a current supply period for supplying a power supply current from a DC power supply to the stator coil of a 3-phase variable speed motor and a freewheeling period for supplying a freewheeling current to the stator coil. In the inverter drive method in which the three-phase inverter is controlled by the pulse width modulation (PWM) method and the winding switching of the stator coil is not performed.
The three-phase inverter consists of three H bridges that form the stator coil and separately supply phase currents to the three phase coils that are independent of each other.
A method for driving an inverter, wherein the current supply periods of the three H bridges do not overlap each other under operating conditions in which the power supply current is less than a predetermined value.

Images