JP7347111B2 - Power converter control method and power converter control device - Google Patents

Description

The present invention relates to a power converter control method and a power converter control device.

A power converter has a plurality of switching elements, and converts DC power into AC power by operating these switching elements. In operating the switching element, the pulse width of the on or off period is changed. Such pulse width control is generally referred to as PWM (Pulse Width Modulation) control. In PWM control, the output from the power converter is controlled by changing the pulse pattern at a predetermined period.

For example, Patent Document 1 shows an example in which the load is a motor, and when performing PWM control to drive the motor, sine wave control, overmodulation control, and , a technique for switching the control mode between square wave control and rectangular wave control has been disclosed. Among these control modes, in the sine wave control and overmodulation control, PWM control is performed, and the switching elements are operated using a pulse pattern according to the operating state of the motor.

Japanese Patent Application Publication No. 2010-213485

Here, if the switching interval in the switching element of the power converter becomes short, a surge voltage may be generated, which may reduce the durability of the power converter. In contrast, in the technique disclosed in Patent Document 1, a pre-stored pulse pattern can be designed so as not to include narrow pulses with extremely short pulse widths. However, even when the pulse pattern is designed in this way, when switching the pulse pattern, the switching interval becomes short before and after the switching, and there is a possibility that narrow pulses may occur.

The present invention has been made to solve the above problems, and provides a power converter control method and a power converter control device that can suppress the generation of narrow pulses when the pulse pattern is switched. do.

The power converter control method of the present invention is a power converter that converts DC power into AC power by changing the conduction state of a switching element using PWM control according to a pulse pattern that is repeated at a predetermined control cycle. This is a method of controlling the device. The power converter control method includes a switching determination step in which it is determined whether or not to switch the pulse pattern according to a control command value for a load connected to the power converter, and a switching determination step in which it is determined whether the pulse pattern is to be switched. In this case, the time between the change timing of the conduction state in the first pulse pattern immediately before the switching timing and the change timing of the conduction state in the second pulse pattern immediately after the switching timing is calculated as follows: a pulse width estimation step in which the pulse width is estimated as a pulse width; a determination step in which it is determined whether the pulse width estimated in the pulse width estimation step exceeds a threshold; and a determination step in which it is determined that the pulse width exceeds the threshold in the determination step. In this case, the method includes a switching step of switching the pulse pattern at the switching timing.

According to the motor control method of the present invention, generation of narrow pulses can be suppressed.

FIG. 1 is a schematic configuration diagram of a motor system including a power converter control device according to a first embodiment. FIG. 2 is a schematic configuration diagram of the control device. FIG. 3 is a schematic configuration diagram of the PWM output section. FIG. 4 is a schematic configuration diagram of the PWM pattern switching output section. FIG. 5 is a flowchart of final command pulse pattern determination control. FIG. 6 is a diagram showing an example of switching command pulse patterns. FIG. 7 is a diagram showing another example of switching command pulse patterns. FIG. 8 is a graph showing the relationship between pulse width and surge voltage. FIG. 9 is a schematic configuration diagram of a PWM pattern switching output section according to the second embodiment. FIG. 10 is a flowchart of final command pulse pattern determination control. FIG. 11 is a diagram showing an example of switching command pulse patterns.

EMBODIMENT OF THE INVENTION Hereinafter, the control method of the power converter and the control apparatus of a power converter which concern on embodiment of this invention are demonstrated using drawing.

(First embodiment)
FIG. 1 is a schematic configuration diagram of a motor system including a power converter and a control device thereof according to a first embodiment.

Power converter 1 is an inverter configured with a plurality of switching elements, and is connected between a positive electrode (P) line and a negative electrode (N) line of battery 2. Power converter 1 converts DC power supplied from battery 2 into AC power and supplies it to motor 3.

In the power converter 1, switching elements are operated according to a command pulse pattern output from the control device 10. Then, the DC power output from the battery 2 is converted into AC power according to the ON/OFF operation of the switching element, and the converted AC power is supplied to the motor 3.

In this embodiment, the motor 3 is driven in three phases, and the power converter 1 includes six switching elements so that it can be driven in three phases and six arms. Note that the configurations of the power converter 1 and the motor 3 are merely examples, and are not limited to the examples of this embodiment. Further, the load connected to the power converter 1 is not limited to the motor 3, but may be other equipment driven by AC power, such as an audio amplifier or a radio.

Capacitor 4 is provided between the positive electrode line and the negative electrode line, and smoothes ripples caused by operation of the switching elements in power converter 1. Further, a voltage sensor 5 is provided at both ends of the capacitor 4, and when the voltage sensor 5 detects a DC voltage V _dc that is a voltage difference between the positive electrode line and the negative electrode line, the DC voltage V _dc is sent to the control device. Output to 10. The DC voltage V _dc is input power to the power converter 1 and is used for control of the power converter 1 by the control device 10 .

Further, a current sensor 6 is provided between the power converter 1 and the motor 3. When the current sensor 6 measures the current measurement values i _u , i _v , i _w of the uvw phase supplied to the motor 3 , it outputs the current measurement values i _u , i _v , i _w to the control device 10 .

The motor 3 is provided with a resolver 7, and using a magnetic flux signal output from the excited resolver 7, a motor rotation angle detection section included in the control device 10 determines the rotor position θe of the motor 3 and Calculate the rotation speed N.

Control device 10 controls the entire motor system 100 including power converter 1. The control device 10 is configured so that a predetermined program can be executed by a microcomputer equipped with a central processing unit (CPU), a read-only memory (ROM), a random access memory (RAM), and an input/output interface (I/O interface). be done. Note that the control device 10 may be composed of a plurality of microcomputers.

In order to perform PWM control that changes the pulse width, the control device 10 generates a command pulse pattern that provides a desired pulse width based on the torque command value T ^* and the rotation speed N of the motor 3, and uses the generated command The pulse pattern is output to the power converter 1. In this way, the switching elements included in the power converter 1 are controlled.

FIG. 2 is a block diagram showing the detailed configuration of the control device 10.

The control device 10 includes a current command value calculation section 11, a current control section 12, a modulation rate conversion section 13, a PWM output section 14, and a three-phase/two-phase conversion section 15. The configuration from the current command value calculation unit 11 to the three-phase/two-phase conversion unit 15 may be logically arranged inside the control device 10, or may be configured by a different microcomputer.

The current command value calculation section 11 receives the torque command value T ^* from the host device and the rotation speed N of the motor 3 from the motor rotation angle detection section of the control device 10. Based on these inputs, the current command value calculation unit 11 calculates current command values i _d ^* , i _q ^* for the motor 3 using a map stored in advance, and calculates current command values i _d ^* , i _q ^*. is output to the current control section 12.

The current control unit 12 receives current command values i _d ^* , i _q ^* output from the current command value calculation unit 11 and current measurement values i _d , i output from the three-phase/two-phase conversion unit 15 (described later). Accepts _q . The current control unit 12 determines the voltage command values V _d ^* , V _q ^* so that the deviation between the current command values _i d *, i _q ^* and the measured ^current values i _d , i _q becomes zero. The determined voltage command values V _d ^* and V _q ^* are output to the modulation rate conversion section 13 .

The modulation rate converter 13 converts the DC voltage V _dc output from the voltage sensor 5 , the rotor position θ _e and the rotation speed N output from the motor rotation angle detector, and the voltage command output from the current controller 12 . The values V _d ^* and V _q ^* are accepted as input. Modulation rate converter 13 calculates modulation rate M ^* of AC power output from power converter 1 based on these inputs.

The PWM output unit 14 receives as input the rotor position θ _e and the rotation speed N output from the motor rotation angle detection unit, and the modulation rate M ^* output from the modulation rate conversion unit 13 . Based on these inputs, the PWM output unit 14 selects and outputs an appropriate pulse pattern from pre-stored pulse patterns used for PWM control. Further, the selection of the pulse pattern is performed every predetermined control period T. In this embodiment, as shown in FIG. 3, which will be described later, the pulse pattern switching timing is controlled in the PWM pattern switching output section 142 included in the PWM output section 14. By controlling the command pulse pattern in this manner, generation of narrow pulses can be suppressed.

The three-phase/two-phase converter 15 uses the rotor position θ _e output from the motor rotation angle detector to convert the uvw phase current measurements i _u , i _v , i _w acquired by the current sensor 6 into On the other hand, phase conversion processing is performed to generate current measurement values i _d and i _q . The three-phase/two-phase conversion unit 15 then outputs the current measurement values i _d and i _q obtained by the conversion process to the current control unit 12 .

FIG. 3 is a block diagram of the detailed configuration of the PWM output section 14. The PWM output section 14 includes a PWM generation section 141 and a PWM pattern switching output section 142.

When the PWM generation unit 141 receives the modulation rate M ^* output from the modulation rate conversion unit 13 and the rotation speed N output from the motor rotation angle detection unit, the PWM generation unit 141 refers to a pre-stored table and performs modulation. The optimum control mode according to the ratio M ^* and the rotation speed N, and the command pulse pattern corresponding to the control mode are determined. Note that the control mode and the command pulse pattern are determined at every predetermined control cycle T, and the command pulse pattern determined at a certain timing is used in the next control cycle T that starts after the control cycle T. .

The PWM generation unit 141 selects either the rectangular wave control mode or the PWM control mode as the control mode. When the rectangular wave control mode is selected, the PWM generation unit 141 outputs, in a certain control period T, a rectangular wave pulse that is one pulse during the entire period of the control period T as a command pulse pattern. When the PWM control mode is selected, the PWM generation unit 141 selects and outputs a pulse pattern used for PWM control according to the modulation rate M ^* and the rotation speed N.

When the square wave control mode is selected, the PWM pattern switching output unit 142 receives the square wave pulse from the PWM generation unit 141 and outputs the square wave pulse as the final command pulse pattern in the next control cycle T. .

On the other hand, when the PWM control mode is selected, upon receiving a predetermined command pulse pattern from the PWM generation section 141, the PWM pattern switching output section 142 outputs the command pulse pattern currently being output and the command pulse pattern received from the PWM generation section 141. It is determined whether the command pulse pattern in the next control cycle T is the same.

If both command pulse patterns are the same, the PWM pattern switching output section 142 continues to output the same command pulse pattern. On the other hand, when the command pulse patterns are different, as will be described later, the PWM pattern switching output unit 142 switches the command pulse pattern using the rotation speed N and the rotor position θ _e so that narrow pulses are not generated. Decide on timing. Then, the PWM pattern switching output unit 142 outputs the final command pulse pattern so that the command pulse pattern can be switched to a new command pulse pattern at the determined switching timing.

FIG. 4 is a block diagram of the detailed configuration of the PWM pattern switching output section 142. The PWM pattern switching output section 142 includes an angle estimation section 1421 and a final command output section 1422.

Here, if the pulse width (pulse width) when PWM control is performed is a narrow pulse that is less than the pulse threshold _Tth , the interval between switching operations performed at the start timing and end timing of the narrow pulse is becomes short and a surge voltage occurs. In the pulse pattern prepared in advance, the pulse width is set to exceed the pulse threshold T _th .

However, before and after the switching timing, the pulse width may become shorter than the pulse threshold T _th . In other words, a pulse that starts at the timing of the last conduction state change of the pulse pattern before the switching timing and ends at the first conduction state change timing of the pulse pattern after the switching timing has a pulse width that is less than the pulse threshold T _th . It is not compensated that it will be too long. Therefore, by performing the following control, the pulse width of the pulse that straddles the switching timing can be made longer than the pulse threshold value T _th , so that the generation of narrow pulses can be suppressed.

Further, the PWM control used to control the motor 3 in this embodiment is called angle-synchronized PWM control. In the command pulse pattern in angle-synchronized PWM control, the rotor position θ _e is associated with the conduction state (on/off). Therefore, even if the switching timing of the command pulse pattern is changed, the final command output unit 1422 outputs a level indicating the conduction state shown in the command pulse pattern according to the rotor position θ _e . .

First, upon receiving the input of the rotor position θ _e and the rotation speed N of the motor 3, the angle estimating unit 1421 calculates the control period T corresponding to the previously stored command pulse pattern and Using the number of pole pairs P of the motor 3, the estimated electrical angle θ _est is calculated by the following equation.

The estimated electrical angle θ _est indicates the switching timing from the current command pulse pattern n to the next command pulse pattern n+1 using the electrical angle of the motor 3. The angle estimation unit 1421 calculates the estimated electrical angle θ _est at a predetermined period, and at a certain switching timing, calculates the estimated electrical angle θ _est indicating the next switching timing.

Therefore, as shown in equation (1), the estimated electrical angle θ _est , which is the next switching timing, is calculated by adding the electrical angle corresponding to the control period T to the current rotor position θ _e . . Note that in equation (1), the control period T is multiplied by "(N/60)·P·360" in order to convert the unit system from time to electrical angle.

Then, the final command output unit 1422 uses the estimated electrical angle θ _est output from the angle estimation unit 1421 to generate a command pulse output from the PWM generation unit 141 from the command pulse pattern n currently being output, according to the following equation. When switching to pattern n+1, the pulse width T _pwm of the pulse that straddles the switching timing is calculated.

However, the parameters in equation (2) are as follows.
θ ₁ : Electrical angle at which conduction state changes last in command pulse pattern n θ ₂ : Electrical angle at which conduction state changes first in command pulse pattern n+1

The electrical angle θ ₁ is an electrical angle that indicates the timing at which the conduction state of the switching element changes at the end in the command pulse pattern n before the estimated electrical angle θ _est which is the switching timing. The electrical angle θ ₂ is an electrical angle indicating the timing at which the conduction state of the first switching element in the command pulse pattern n+1 changes after the estimated electrical angle θ _est .

That is, (θ _est - _{θ 1} ) is the pulse width before the estimated electrical angle θ _est expressed in electrical angles, and (θ ₂ - θ _est ) is the pulse width after the estimated electrical angle θ _est expressed in electrical angles. It is shown by the corner. Therefore, in order to convert the unit system from electrical angle to time for the sum of the two, "(θ _est - _{θ 1} ) + (θ ₂ - θ _est )", we need to convert the unit system from electrical angle to time by "360・(N/60 )·P”, the pulse width T _pwm of the pulse that straddles the switching timing can be obtained as shown in equation (2).

In this example, it is assumed that the conduction state of the command pulse pattern n and the command pulse pattern n+1 is the same at the estimated electrical angle θ _est . Therefore, the conduction state is switched on at electrical angle θ ₁ and switched off at electrical angle θ ₂ , and as a result, the electrical angle θ ₁ and electrical angle θ ₂ before and after the estimated electrical angle θ _est are switched. The pulse straddles the timing and becomes an on period. Note that an example in which the command pulse pattern n and the command pulse pattern n+1 are different in the estimated electrical angle θ _est will be described in the second embodiment.

Then, if the pulse width T _pwm is equal to or greater than the pulse threshold T _th , the final command output unit 1422 switches the command pulse pattern at the estimated electrical angle θ _est . On the other hand, when the pulse width T _pwm is shorter than the pulse threshold T _th , the angle estimation unit 1421 prevents the pulse width T _pwm from becoming shorter than the pulse threshold T _th in order to suppress the generation of narrow pulses. Reset the switching timing. This resetting of the switching timing and the final output control of the command pulse pattern will be explained using FIG. 5.

FIG. 5 is a flowchart of final command pulse pattern determination control. In the control device 10, this decision control is repeatedly performed at predetermined timing. In the following, for clarity of explanation, the blocks that implement predetermined functions in the control device 10 shown in FIGS. 2 to 4 will be described as performing respective processes. Note that a predetermined function may be realized by the control device 10 performing a predetermined process.

First, in step S101, the PWM generation unit 141 determines whether PWM control is being performed. If the PWM generation unit 141 is performing PWM control (S101: Yes), then it performs the process of step S103. The PWM generation unit 141 performs rectangular wave control, and if it does not perform PWM control (S101: No), then it performs the process of step S102.

In step S102, the PWM generation unit 141 outputs a pulse pattern in the selected control mode, rectangular wave control, as a final command pulse pattern. Note that when rectangular wave control is performed, the PWM generation unit 141 outputs an ON or OFF conduction state so that one pulse is generated over the entire control period T.

In step S103, the modulation rate conversion unit 13 converts the DC voltage V _dc , the rotor position θ _e and the rotation speed N output from the motor rotation angle detection unit, and the voltage command value V _d output from the current control unit 12. ^* , V _q ^* are accepted, and the modulation factor M ^* is calculated based on these inputs. Note that the rotation speed N may be calculated by the modulation rate conversion unit 13 differentiating the rotor position θ _e with respect to time. Then, the process of step S104 is performed.

In step S104, the PWM generation unit 141 determines a command pulse pattern based on the modulation rate M ^* and the rotation speed N generated in step S102. Then, the final command output unit 1422 determines whether the current command pulse pattern being output is the same as the determined next command pulse pattern. If both are the same and no change is necessary (S104: No), then the process of step S105 is performed. If the two are different and it is necessary to change the command pulse pattern to be output (S104: Yes), then the process of step S106 is performed.

In step S105, the final command output unit 1422 outputs the same command pulse pattern as the currently output command pulse pattern in the next control cycle T without changing the command pulse pattern. In this way, the final command pulse pattern is determined.

In step S106, upon receiving the input of the rotor position θ _e and the rotation speed N, the angle estimating unit 1421 calculates these input values, the control period T corresponding to the pre-stored pulse pattern, and Using the number P of pole pairs of the motor 3, the pulse pattern switching timing is calculated as an estimated electrical angle θ _est expressed in electrical angles using equation (1). This estimated electrical angle θ _est indicates the timing when switching is performed every predetermined control period T. Once the estimated electrical angle θ _est is calculated, the process of step S107 is then performed.

In step S107, the angle estimation unit 1421 determines the electrical angle θ ₁ at which the conduction state is switched in the command pulse pattern n before the estimated electrical angle θ _est , and the electrical angle at which the conduction state is switched in the command pulse pattern n+1 after the estimated electrical angle θ _est . Using θ ₂ , the rotational speed N and the number of pole pairs P of the motor 3, the pulse width T _pwm in the case where the command pulse pattern is switched at the estimated electrical angle θ _est is calculated. Once the pulse width T _pwm has been calculated, the process of step S108 is then performed.

In step S108, the final command output unit 1422 determines whether the pulse width T _pwm calculated in step S106 is greater than or equal to a predetermined pulse threshold T _th . If the pulse width T _pwm is greater than or equal to the pulse threshold value T _th (S108: Yes), the final command output unit 1422 determines that the pulse width T _pwm is sufficiently long and no narrow pulse is generated, and then performs the step The process of S109 is executed. If the pulse width T _pwm is short and not equal to or greater than the pulse threshold T _th (S108: No), the final command output unit 1422 determines that there is a possibility that a narrow pulse will occur, and then performs the process of step S110. Execute.

In step S109, the final command output unit 1422 determines the estimated electrical angle θ _est as the switching timing of the command pulse pattern, outputs the command pulse pattern n before the estimated electrical angle θ _est , and outputs the command pulse pattern n after the estimated electrical angle θ _est . In this case, a command pulse pattern n+1 is output. In this way, the final command pulse pattern is determined so that the pulse pattern switches at the estimated electrical angle θ _est .

In step S110, the angle estimating unit 1421 determines the electrical angle after the estimated electrical angle θ _est at which the conduction state in the command pulse pattern n and the conduction state in the command pulse pattern n+1 are the same as the command pulse It is reset as the estimated electrical angle θ _est , which is the pattern switching timing. Since the command pulse pattern can be switched at the timing when the conduction state of the command pulse pattern n and the command pulse pattern n+1 match, the timing is reset as the estimated electrical angle θ _est . Note that this resetting process may be referred to as a first resetting process to distinguish it from the resetting process in the second embodiment.

Then, the process of step S107 is executed again using the estimated electrical angle θ _est reset in step S110. Such resetting of the estimated electrical angle θ _est is repeated until the pulse width T _pwm becomes equal to or greater than the pulse threshold value T _th (S108: Yes).

Next, output control of the command pulse pattern in this embodiment will be explained using FIGS. 6 and 7. These diagrams show, from the top, the command pulse pattern n currently being output, the command pulse pattern n+1 in the next control cycle T generated by the PWM generation unit 141, and the command pulse pattern n+1 output from the final command output unit 1422. The final command pulse pattern is shown. Note that the current command pulse pattern n before the switching timing is shown by a solid line, and after the switching timing is shown by a dotted line. Further, in the next command pulse pattern n+1, the period before the switching timing is shown by a dotted line, and the period after the switching timing is shown by a solid line.

In FIG. 6, the estimated electrical angle θ _est calculated in step S106 is the switching timing of the command pulse pattern, and in FIG. 7, the estimated electrical angle θ _est reset in step S110 is the switching timing of the command pulse pattern. shall become.

First, the example shown in FIG. 6 will be explained.

In FIG. 6, the estimated electrical angle θ _est calculated in step S106 is shown. The upper row shows the electrical angle θ ₁ at which the last conduction state switches in the command pulse pattern n before the estimated electrical angle θ _est , and the middle row shows the electrical angle θ 1 at which the last conduction state switches in the command pulse pattern n+1 after the estimated electrical angle θ _est . The electrical angle θ ₂ at which the initial conduction state switches is shown.

Through the process of step S107, a pulse width T _pwm corresponding to these electrical angles θ ₁ and θ ₂ is calculated. Here, since the pulse width T _pwm is greater than or equal to the pulse threshold T _th (S108: Yes), the command pulse pattern is changed at the estimated electrical angle θ _est , as shown in the lower part of the figure.

Next, the example shown in FIG. 7 will be explained.

In FIG. 7, the estimated electrical angle θ _est calculated in step S106 and the estimated electrical angle θ _est reset in step S110 are shown in the upper row.

In the upper row, the electrical angle θ ₁ at which the conduction state is last switched in the command pulse pattern n before the estimated electrical angle θ _est is shown. In the middle row, the electrical angle θ ₂ at which the first conduction state is switched in the command pulse pattern n+1 after the first calculated estimated electrical angle θ _est , and the command pulse pattern after the estimated electrical angle θ _est to be reset. The electrical angle θ ₂ at which the conduction state is first switched at n+1 is shown. In the lower row, the final command pulse pattern that is finally output is shown.

First, in the process of step S107, a pulse width T _pwm is calculated using the first calculated estimated electrical angle θ _est and the corresponding electrical angles θ ₁ and θ ₂ , and this pulse width T _pwm is equal to the pulse threshold T _th shorter (S108: No). Therefore, it is determined that there is a possibility that a narrow pulse will occur, and then the process of step S110 is executed. In step S110, the angle estimating unit 1421 determines the conduction state (in this figure, the on state) in the command pulse pattern n that is being output after the first calculated estimated electrical angle θ _est , and The electrical angle at which the conduction state (on state) is the same in command pulse pattern n+1 is reset as the estimated electrical angle θ _est .

Then, by the process of step S107 again, the pulse width T _pwm is calculated according to the electrical angles θ ₁ and θ ₂ corresponding to the estimated electrical angle θ _est to be reset. Here, since the pulse width T _pwm is greater than or equal to the pulse threshold T _th (S108: Yes), the command pulse pattern is switched at the recalculated estimated electrical angle θ _est , as shown in the lower part of the figure. Therefore, the pulse width T _pwm of the pulse that straddles the switching timing of the command pulse pattern becomes equal to or greater than the pulse threshold value T _th , so that the generation of narrow pulses can be suppressed.

Note that FIG. 8 is a graph showing the relationship between the pulse width T _pwm and the generated surge voltage. In this graph, the pulse width T _pwm is shown in the x-axis direction, and the generated surge voltage is shown in the y-axis direction. Also, according to this figure, the characteristics change depending on the temperature of the power converter 1, and the characteristics when the power converter 1 is at 40 degrees Celsius, 20 degrees Celsius, 0 degrees Celsius, and -40 degrees Celsius are shown by solid lines. , indicated by a dash-dotted line, a dash-double-dot line, and a dashed line.

For example, by setting the pulse threshold value T _th to 1.65 μs, the pulse width T _pwm becomes 1.65 μs or more, and as a result, it is possible to prevent a surge voltage from occurring at any of these temperatures. In this way, the criteria for determination in step S108 can be set in advance using the characteristics of the surge voltage depending on the temperature of the power converter 1.

According to the first embodiment, the following effects can be obtained.

According to the control method for the power converter 1 of the first embodiment, it is determined whether or not to switch the command pulse pattern according to the torque command value T ^* for the motor 3, which is a load connected to the power converter 1. a switching determination step (S104), and a pulse width estimation step (S107) of estimating a pulse width T _pwm that straddles the estimated electrical angle θ _est , which is the switching timing, when it is determined in the switching determination step that the command pulse pattern is to be switched. , a determination step (S108) in which it is determined whether the pulse width T _pwm estimated in the pulse width estimation step exceeds the pulse threshold T _th , and it is determined that the pulse width T _pwm exceeds the pulse threshold T _th If (S108: Yes), a switching step (S109) of switching the command pulse pattern at the estimated electrical angle θ _est is included.

With this configuration, the command pulse pattern is switched only when the pulse width T _pwm is longer than the pulse threshold T _th , so even under transient conditions when the command pulse pattern is switched. , the generation of narrow pulses can be prevented.

Furthermore, in the pulse width estimation step (S107), the operation timing (θ ₁ ) of the command pulse pattern n immediately before the estimated electrical angle θ _est , which is the switching timing, and the operation of the command pulse pattern n+1 immediately after the estimated electrical angle θ _est are determined. The time corresponding to the timing (θ ₂ ) is estimated as the pulse width T _pwm . Thereby, the pulse width T _pwm is estimated using the estimated electrical angle θ _est , the command pulse pattern n, and the command pulse pattern n+1. In this way, there is no need to add a new sensor for resetting the estimated electrical angle θ _est , which is the switching timing, so the generation of narrow pulses can be prevented without increasing the processing load.

According to the control method for the power converter 1 of the first _embodiment , _the first _resetting step (S110 ).

With this configuration, when the pulse width T _pwm is shorter than the pulse threshold T _th , the estimated electrical angle θ _est , which is the switching timing, is reset, and the pulse width T _pwm becomes the pulse threshold T _th . exceed. Therefore, generation of narrow pulses can be prevented even under transient conditions where command pulse patterns are switched.

According to the control method for the power converter 1 of the first embodiment, in the resetting step (S110), after the estimated electrical angle θ _est calculated in step S106, the first pulse pattern is The timing at which the conduction state becomes the same as the second pulse pattern is reset as the estimated electrical angle θ _est .

If the estimated electrical angle θ _est is set to the timing at which the conduction state in the command pulse pattern changes after the estimated electrical angle θ _est , which is the switching timing, a change in the conduction state will occur at the estimated electrical angle θ _est , and the pulse width T _pwm may fall below the pulse threshold T _th . However, by resetting the timing at which the conduction state becomes the same as the estimated electrical angle θ _est , the pulse width T _pwm can be made to exceed the pulse threshold value T _th , thereby preventing the generation of narrow pulses.

According to the control method for the power converter 1 of the first embodiment, the load connected to the power converter 1 is the motor 3, and the PWM generation unit 141 generates a torque command value T ^* for the motor 3 and a torque command value T* for the motor 3. A pulse pattern is generated based on the rotation speed N of. With such a configuration, even when the load is the motor 3, generation of narrow pulses can be suppressed.

According to the method for controlling the power converter 1 of the first embodiment, the power converter 1 outputs AC power to the motor 3. As shown in FIG. 3, in the determination step (S104) performed by the PWM generation unit ¹⁴¹ , the command pulse pattern is It is determined whether or not to switch.

Further, in the pulse width estimation step (S106), the electrical angle θ ₁ of the motor 3 indicating the timing at which the conduction state of the first pulse pattern immediately before the estimated electrical angle θ _est is switched, and the second electrical angle immediately after the estimated electrical angle θ _est are determined. The pulse width T _pwm is estimated according to the electrical angle θ ₂ of the motor 3, which indicates the timing at which the conductive state is switched at the beginning of the pulse pattern, and the rotational speed N of the motor 3.

By being controlled in this way, the power converter 1 is controlled so that the pulse width T _pwm exceeds the pulse threshold value T _th based on the parameters related to the motor 3 to which AC power is output, thereby improving the design accuracy. It is possible to prevent the generation of narrow pulses while increasing the

According to the control method for the power converter 1 of the first embodiment, the pulse threshold value T _th is determined by the surge voltage characteristic according to the temperature of the switching element included in the power converter 1. Thereby, the pulse threshold value T _th can be determined in advance, so that generation of narrow pulses can be prevented depending on the operating environment.

(Second embodiment)
In the first embodiment, an example has been described in which the rotation speed N calculated in the motor rotation angle detection section is used when calculating the estimated electrical angle θ _est . In the second embodiment, an example of estimating the rotation speed N of the motor 3 at the timing when the command pulse pattern changes when the rotation speed N of the motor 3 changes will be described.

FIG. 9 is a schematic configuration diagram of the PWM pattern switching output section 142 of the second embodiment. Compared to the PWM pattern switching output section 142 of the first embodiment shown in FIG. 3, a rotation speed estimating section 901 is provided before the angle estimating section 1421.

The rotation speed estimating unit 901 stores the rotation speed N for each control period T that is a switching timing. Note that in the following, the rotation speed N _n indicates the rotation speed of the motor 3 in the n-th control cycle T.

The rotation speed estimation unit 901 estimates the estimated rotation speed N _est , which is the estimated value of the rotation speed N _n+1 at the next switching timing, using the following equation.

However, the parameters in equation (3) are as follows.
N _n-1 : Number of revolutions N at the previous (n-1st) switching timing
N _n : Number of revolutions N at the current (nth) switching timing
N _est : Estimated value of rotation speed N at the next (n+1th) switching timing

According to equation (3), it is assumed that the amount of change in the rotational speed N from the previous switching timing to the current switching timing is equal to the amount of change in the rotational speed N from the current switching timing to the next switching timing. Then, an estimated value N _n+1 of the rotation speed N at the next switching timing is estimated.

FIG. 10 is a flowchart of final command pulse pattern determination control. Compared to the flowchart of the first embodiment shown in FIG. 5, the process of step S106 is changed to step S202, and the processes of steps S201, S203, and S204 are added.

Step S201 is a process performed when it is necessary to change the command pulse pattern to be output (S104: Yes). In this process, the rotation speed estimation unit 901 calculates the estimated rotation speed N _est at the next change timing based on equation (3). Then, the process of step S202 is performed.

In step S202, the angle estimating unit 1421 compares the current rotation speed N and the estimated rotation speed N _est , and uses the larger rotation speed to calculate the estimated electrical angle θ _est using equation (1). With this control, even when the rotational speed N changes greatly from a rapidly decreasing state to a steady state, a larger rotational speed N is used, so that the pulse width T _pwm becomes shorter. Therefore, it is difficult to judge that the pulse width T _pwm is equal to or greater than the pulse threshold value T _th , so it is possible to suppress the possibility that the pulse width T _pwm becomes a narrow pulse.

Note that in step S202, the angle estimating unit 1421 may calculate the estimated electrical angle θ _est using equation (1) using the estimated rotational speed N _est . By doing so, the processing load can be reduced.

In step S203, the angle estimation unit 1421 determines whether the conduction state in the previous command pulse pattern and the conduction state in the next command pulse pattern are the same at the estimated electrical angle θ _est . If both are the same (S203: Yes), then step S107 is performed. On the other hand, if the two are different (S203: No), then step S204 is performed.

In step S204, the angle estimation unit 1421 newly estimates an electrical angle in which the conduction state of the current command pulse pattern n and the conduction state of the next command pulse pattern n+1 match after the estimated electrical angle θ _est . Reset as electrical angle θ _est . Note that this resetting process may be referred to as a second resetting process to distinguish it from the resetting process (first resetting process) in step S110.

Note that the process corresponding to step S204 may be included in the control that does not include the process of estimating the rotation speed N, as in the first embodiment. By doing so, the conduction state between the current command pulse pattern n and the next command pulse pattern n can be made the same at the estimated electrical angle θ _est , so that generation of narrow pulses can be suppressed.

Next, output control of the command pulse pattern in this embodiment will be explained using FIG. 11. These diagrams show, from the top, the command pulse pattern n currently being output, the command pulse pattern n+1 in the next control cycle T generated by the PWM generation unit 141, and the command pulse pattern n+1 output from the final command output unit 1422. The final command pulse pattern is shown.

In FIG. 11, the estimated electrical angle θ _est calculated in step S204 is the switching timing of the command pulse pattern.

As shown in this figure, in the estimated electrical angle θ _est first calculated in step S106, the conduction state of command pulse pattern n and the conduction state of command pulse pattern n+1 are different (S203: No), so , after the estimated electrical angle θ _est , the timing at which the conduction state of command pulse pattern n and the conduction state of command pulse pattern n+1 are the same is reset as a new estimated electrical angle θ _est (S204). .

In this case, as shown in the upper row, the electrical angle θ ₁ at which the conduction state of the command pulse pattern n is switched immediately before the estimated electrical angle θ _est after the reset is the estimated electrical angle θ _est after the reset. matches. Further, in the middle stage, an electrical angle θ ₂ at which the conduction state of the command pulse _pattern n+1 is switched is shown immediately after the estimated electrical angle θ est after resetting.

Through the process of step S107, a pulse width T _pwm corresponding to these electrical angles θ ₁ and θ ₂ is calculated. Here, since the pulse width T _pwm is greater than or equal to the pulse threshold T _th (S108: Yes), the command pulse pattern is changed at the estimated electrical angle θ _est , as shown in the lower part of the figure. Therefore, it is possible to suppress the pulse width T _pwm that straddles the switching timing of the command pulse pattern from becoming a narrow pulse.

According to the second embodiment, the following effects can be obtained.

According to the control method for the power converter 1 of the second embodiment, the method further includes a rotation speed estimation step (S201) of estimating the rotation speed N at the switching timing based on the change in the rotation speed N for each control period T. . In the pulse width estimation step (S106), the pulse width T _pwm is estimated using the estimated rotation speed N _est estimated in the rotation speed estimation step. With such a configuration, generation of narrow pulses can be prevented even when the rotational speed N of the motor 3 is rapidly increasing.

According to the control method for the power converter 1 of the second embodiment, in the pulse width estimation step (S202), the estimated rotation speed N _est estimated in the rotation speed estimation step (S201) and the current rotation speed N The pulse width T _pwm is estimated using the larger of the two. With this configuration, the pulse width T _pwm is calculated to be shorter, making it difficult to judge that the pulse width T _pwm is equal to or greater than the pulse threshold T _th , reducing the possibility that the pulse width T _pwm becomes a narrow pulse. This can be suppressed.

According to the control method for the power converter 1 of the second embodiment, at the estimated electrical angle θ _est calculated using the estimated rotation speed N _est estimated in the rotation speed estimation step (S201), If the pattern and the next command pulse pattern are different (S203: No), the electrical angle at which the conduction state matches between the current command pulse pattern and the next command pulse pattern is set as the estimated electrical angle θ _est . 2 resetting step (S204). With this configuration, by resetting the timing at which the conduction state becomes the same as the estimated electrical angle θ _est , the pulse width T _pwm exceeds the pulse threshold T _th and the generation of narrow pulses is prevented. be able to.

Although the embodiments of the present invention have been described above, the above embodiments merely show a part of the application examples of the present invention, and are not intended to limit the technical scope of the present invention to the specific configurations of the above embodiments. do not have. Furthermore, the above embodiments can be combined as appropriate.

1 Power converter 3 Motor 6 Current sensor 10 Control device 11 Current command value calculation unit 12 Current control unit 13 Modulation rate conversion unit 14 PWM output unit 141 PWM generation unit 142 PWM pattern switching output unit 1421 Angle estimation unit 1422 Final command output unit 100 Motor system 901 Rotation speed estimation section

Claims (10)

A power converter control method for converting DC power into AC power by changing the conduction state of a switching element using PWM control according to a pulse pattern repeated at a predetermined control cycle, the method comprising:
a switching determination step of determining whether to switch the pulse pattern according to a control command value for a load connected to the power converter;
In the switching determination step, when it is determined that the pulse pattern is to be switched, the timing of a change in the conduction state in the first pulse pattern immediately before the switching timing, and the change in the conduction state in the second pulse pattern immediately after the switching timing. a pulse width estimating step of estimating the time between the switching timings as the pulse width of the pulse that straddles the switching timing;
a determining step of determining whether the pulse width estimated in the pulse width estimating step exceeds a threshold;
A method for controlling a power converter, comprising: a switching step of switching the pulse pattern at the switching timing when the pulse width is determined to exceed the threshold in the determining step. A method for controlling a power converter according to claim 1, comprising:
The method further comprises a first resetting step of resetting the switching timing so that the pulse width exceeds the threshold when it is determined in the determining step that the pulse width does not exceed the threshold. How to control the converter. A method for controlling a power converter according to claim 2, comprising:
Control of the power converter, in the first resetting step, resetting a timing at which the first pulse pattern and the second pulse pattern have the same conduction state after the switching timing as the switching timing. Method. A method for controlling a power converter according to any one of claims 1 to 3, comprising:
If the conduction state between the first pulse pattern and the second pulse pattern is different at the switching timing determined in the switching determining step, then the first pulse pattern and the first pulse pattern are switched after the switching timing. A method for controlling a power converter, further comprising a second resetting step of resetting a timing at which the conduction state becomes the same as the second pulse pattern as the switching timing. A method for controlling a power converter according to any one of claims 1 to 4, comprising:
The load is a motor,
A method for controlling a power converter, further comprising a pulse pattern determining step of determining the pulse pattern based on a torque command value for the motor and a rotation speed of the motor. A method for controlling a power converter according to claim 5,
A method for controlling a power converter, wherein in the switching determination step, it is determined whether or not to switch a pulse pattern based on a modulation rate of the AC power with respect to the DC power in the power converter and the rotation speed. The power converter control method according to claim 5 or 6,
further comprising a rotation speed estimating step of estimating the rotation speed at the switching timing based on the change in the rotation speed for each control cycle,
A method for controlling a power converter, wherein in the pulse width estimating step, the estimated rotation speed estimated in the rotation speed estimation step is used as the rotation speed to estimate the pulse width. A method for controlling a power converter according to claim 5 or 6,
further comprising a rotation speed estimating step of estimating the rotation speed at the switching timing based on the change in the rotation speed for each control cycle,
In the pulse width estimating step, the power converter estimates the pulse width using the larger of the estimated rotation speed estimated in the rotation speed estimation step and the current rotation speed as the rotation speed. control method. A method for controlling a power converter according to any one of claims 1 to 8, comprising:
A method for controlling a power converter, wherein a threshold value is determined by characteristics of a surge voltage depending on a temperature of the switching element. A control device for a power converter that converts DC power into AC power by changing the conduction state of a switching element using PWM control according to a pulse pattern repeated at a predetermined control cycle,
The control device includes:
determining whether to switch the pulse pattern according to a control command value for a load connected to the power converter;
When it is determined that the pulse pattern is to be switched, the time between the timing at which the conduction state changes in the first pulse pattern immediately before the switching timing and the timing at which the conduction state changes in the second pulse pattern immediately after the switching timing. is estimated as the pulse width of the pulse that straddles the switching timing,
determining whether the estimated pulse width exceeds a threshold;
A power converter control device that switches the pulse pattern at the switching timing when it is determined that the pulse width exceeds the threshold value.

Images