Drive circuit and power converter

Abstract

To provide a drive circuit in which both noise suppression and switching loss suppression can be achieved.SOLUTION: A drive IC 100 drives a switching element 200 having a gate electrode. The drive IC 100 includes a current output circuit 20 in which a switching speed can be adjusted by control on a gate current to be outputted to the gate electrode, a DP determination unit 11 that sets, as a drive pattern of the current output circuit, an actual drive pattern corresponding to an operation condition of the switching element, and drives the current output circuit in the actual drive pattern, an SWF monitor unit 13 that acquires a monitor value of a surge voltage waveform following the operation of the switching element, and an EN estimation unit 12 that acquires an estimated value of an EMI noise level from the monitor value. The DP determination unit changes the actual drive pattern corresponding to the operation condition so as to adjust an EMI noise to fall within prescribed noise values and adjust a switching loss to fall within prescribed loss values according to deviation of the estimated value from an EMI noise level target value.SELECTED DRAWING: Figure 2

Description

Technical Field

[0001] The present disclosure relates to a drive circuit and a power conversion device.

Background Art

[0002] Conventionally, as an example of a drive circuit, there is a gate drive device disclosed in Patent Document 1. The gate drive device includes a pull-up circuit having a plurality of PMOS transistors connected in parallel, a pull-down circuit having a plurality of NMOS transistors connected in parallel, and the like.

[0003] The PMOS transistors of the pull-up circuit are individually turned on or off by a 63-bit gate signal. The pull-up circuit supplies a gate current corresponding to the number of turned-on transistors among the PMOS transistors to the gate of the IGBT. The NMOS transistors of the pull-down circuit are individually turned on or off by a 63-bit gate signal. The pull-down circuit supplies a gate current corresponding to the number of turned-on transistors among the NMOS transistors to the gate of the IGBT.

Prior Art Documents

Patent Documents

[0004]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0005] In a gate drive circuit, in order to reduce switching loss, the drive patterns of the pull-up circuit and the pull-down circuit may be changed to increase the switching speed of the IGBT. In this case, in the gate drive circuit, there is a risk that the noise generated along with the switching operation of the IGBT increases.

[0006] One objective of the disclosure is to provide a drive circuit that achieves both noise suppression and switching loss suppression. Another objective of the disclosure is to provide a power converter that achieves both noise suppression and switching loss suppression. [Means for solving the problem]

[0007] The drive circuit disclosed herein is A drive circuit for driving a switching element (200) having a gate electrode, By controlling the gate current output to the gate electrode, vinegar A current control unit (20) that can adjust the switching speed, A drive mode setting unit (11) sets the actual drive mode as the drive mode of the current control unit according to the operating conditions of the switching element, and drives the current control unit in the actual drive mode, A monitoring unit (13) acquires a monitor value of the surge voltage waveform associated with the operation of the switching element, It includes a prediction unit (12, 12a) that acquires a predicted value of the noise level, which is the state of noise generated in conjunction with the operation of the switching element, from the monitor value, The drive mode setting unit changes the actual drive mode according to the operating conditions, based on the deviation between the predicted value and the target value of the noise level, so that the noise is within the specified noise limit and the switching loss of the switching element is within the specified loss limit. death, Furthermore, a drive mode storage unit (50) stores drive mode data in which each of multiple operating conditions is associated with a drive mode suitable for each operating condition, The system includes a learning unit (15) that, when the driving mode is changed to a learning driving mode different from the actual driving mode by the driving mode setting unit, updates the driving mode data by associating the learning driving mode with the operating conditions associated with the actual driving mode when the deviation between the predicted value and the target value falls within a specified range, according to the current control unit being driven in the learning driving mode. It is characterized by the following:

[0008] According to the drive circuit disclosed herein, a predicted noise level is obtained from the monitored value of the surge voltage waveform associated with the operation of the switching element. Then, the drive circuit changes the actual driving mode according to the operating conditions so that the noise level is within the specified limit and the loss limit is within the specified limit, depending on the deviation between the predicted value and the target noise level. In this way, the drive circuit can achieve both noise suppression and switching loss suppression.

[0009] The power conversion device disclosed herein is The present invention is characterized by including the above-mentioned drive circuit and a switching element driven by the drive circuit.

[0010] The power converter disclosed herein includes the above-described drive circuit. Therefore, the power converter can achieve both noise suppression and switching loss suppression.

[0011] The various embodiments disclosed in this specification employ different technical means to achieve their respective objectives. The claims and the reference numerals in parentheses in this section are illustrative in their correspondence with the embodiments described later and are not intended to limit the technical scope. The objectives, features, and effects disclosed in this specification will become clearer by referring to the subsequent detailed description and the accompanying drawings. [Brief explanation of the drawing]

[0012] [Figure 1] This is a block diagram showing the schematic configuration of the power converter in the first embodiment. [Figure 2] This is a block diagram showing the schematic configuration of the drive IC in the first embodiment. [Figure 3] This is a flowchart showing the operation of the drive IC in the first embodiment. [Figure 4] This block diagram shows the schematic configuration of the drive IC in the second embodiment. [Figure 5] This is a flowchart showing the operation of the drive IC in the second embodiment. [Modes for carrying out the invention]

[0013] In the following, several embodiments for implementing this disclosure will be described with reference to the drawings. In each embodiment, parts corresponding to matters described in a previous embodiment may be denoted by the same reference numerals, and redundant explanations may be omitted. If only a part of the configuration is described in each embodiment, the other parts of the configuration can be referred to and applied to other embodiments described in advance.

[0014] (First Embodiment) The power conversion device 1 and the drive IC 100 will be described with reference to FIGS. 1 to 3.

[0015] <Power Conversion Device> As shown in FIG. 1, the power conversion device 1 includes a plurality of drive ICs 100 and a plurality of switching elements 200. The power conversion device 1 is electrically connected to a battery 400 and a motor (rotating electrical machine) 300. Note that the drive IC 100 corresponds to a drive circuit. The motor 300 can employ a motor generator.

[0016] The motor 300 is a three-phase motor including a stator and a rotor. The stator is an electromagnet in which coils 301, 302, and 303 are wound around a core member. Therefore, it can be said that the motor 300 includes U-phase coil 301, V-phase coil 302, and W-phase coil 303. The battery 400 is a DC power source. Also, a smoothing capacitor 500 is provided in parallel with the battery 400. The power conversion device 1 converts the power of the battery 400 into three-phase AC power and supplies it to the motor 300. That is, the power conversion device 1 is a three-phase inverter circuit. The power conversion device 1 can be mounted on a moving body including, for example, a vehicle or an aircraft. Here, as an example, the power conversion device 1 mounted on a vehicle is adopted.

[0017] As the plurality of switching elements 200, the power conversion device 1 includes upper arm elements for each of the U-phase, V-phase, and W-phase, and lower arm elements for each of the U-phase, V-phase, and W-phase. Therefore, the power conversion device 1 includes six switching elements 200. The switching element 200 can adopt a MOSFET or IGBT having a gate electrode. Also, the switching element 200 can adopt one having Si as a main component or one having SiC or the like as a main component. In the present embodiment, as an example of the switching element 200, a SiC-MOSFET is adopted.

[0018] Multiple drive ICs 100 are provided individually for each of the multiple switching elements 200. Therefore, the power converter 1 is equipped with six drive ICs 100. The drive ICs 100 are connected to the gate electrodes of the switching elements 200. The drive ICs 100 drive the switching elements 200. In other words, the drive ICs 100 control the on / off state of the switching elements 200. Therefore, the drive ICs 100 can also be called gate drive circuits or gate drive devices.

[0019] In addition to the components described above, the power converter 1 may also be equipped with various sensors such as a voltage sensor 501, a current sensor 502, and a temperature sensor 503. The voltage sensor 501 detects the voltage of the battery 400 (battery voltage). The current sensor 502 detects the current (phase current) of each coil 301 to 303. In other words, the power converter 1 is equipped with three current sensors 502 to detect the phase currents of the U-phase coil 301, the V-phase coil 302, and the W-phase coil 303, respectively.

[0020] The temperature sensor 503 detects the temperature (element temperature) of the switching element 200. A temperature sensor 503 is provided corresponding to each of the multiple switching elements 200. Therefore, the power converter 1 is equipped with the same number of temperature sensors 503 as the switching elements 200.

[0021] Each of the sensors 501 to 503 is electrically connected to the drive IC 100. Each of the sensors 501 to 503 outputs a sensor signal, which is an electrical signal detected by the sensor, to the drive IC 100.

[0022] <Drive IC> The configuration of the drive IC 100 will be explained using Figure 2. Multiple drive ICs 100 have the same configuration. Therefore, this explanation will use one drive IC 100. In Figure 2, some of the components are shown by abbreviations. DP is an abbreviation for drive pattern. EN is an abbreviation for EMI noise. SWF is an abbreviation for surge voltage waveform. ENC is an abbreviation for EMI noise correlation data. ENT is an abbreviation for EMI noise target value. OC is an abbreviation for operating conditions. Also, EMI is an abbreviation for Electro Magnetic Interference.

[0023] The drive IC 100 includes a computer 10, a current output circuit 20, an ENT memory unit 30, an OC monitor unit 40, a PD memory unit 50, and the like.

[0024] Computer 10 is equipped with processing units such as a CPU, storage devices such as RAM and ROM, and interface circuits such as an AD converter. Computer 10 executes programs stored in the storage devices via its processing units. By executing these programs, Computer 10 performs various arithmetic operations using data stored in RAM and data input via the AD converter. In this way, Computer 10 can achieve various functions. Furthermore, Computer 10 can be said to have multiple functional blocks.

[0025] The computer 10 comprises a DP determination unit 11, an EN prediction unit 12, an SWF monitor unit 13, an ENC memory unit 14, and a learning unit 15 as functional blocks. The DP determination unit 11 corresponds to the drive mode setting unit. The DP determination unit 11 determines the optimal drive pattern as the drive pattern for the current output circuit 20, which will be explained later. The DP determination unit 11 sets (selects) an actual drive pattern as the drive pattern for the current output circuit 20 according to the operating conditions of the switching element 200, etc. The DP determination unit 11 drives the current output circuit 20 with the set actual drive pattern.

[0026] The drive pattern is data indicating the driving modes of the multiple current sources in the current output circuit 20. The switching speed of the current output circuit 20 changes according to the driving modes of the multiple current sources. Therefore, the drive pattern can also be said to be data indicating the switching speed of the switching element 200. Furthermore, the drive pattern can also be said to be data indicating the gate current. Moreover, the DP determination unit 11 can be said to determine the switching speed of the switching element 200.

[0027] The DP determination unit 11 acquires the battery voltage, phase current, and element temperature monitored by the OC monitor unit 40 (which will be explained later) as operating conditions. In other words, the DP determination unit 11 receives the sensor signals from the OC monitor unit 40 as operating conditions for the switching element 200. These operating conditions can also be described as the operating conditions or operating status under which the switching element 200 is actually operating.

[0028] Furthermore, the DP determination unit 11 sets (selects) a drive pattern associated with the operating conditions from among multiple drive patterns stored in the PD memory unit 50, which will be explained later, as the actual drive pattern. The DP determination unit 11 then drives the current output circuit 20 with the set actual drive pattern. The actual drive pattern can also be said to be the optimal drive pattern according to the operating conditions. The actual drive pattern corresponds to the actual drive mode.

[0029] Furthermore, the DP determination unit 11 changes the drive pattern according to the deviation between the predicted EMI noise level and the target EMI noise level. In other words, the DP determination unit 11 changes the actual drive pattern according to the deviation so that the EMI noise is within the specified noise limit and the switching loss of the switching element 200 is within the specified loss limit. The DP determination unit 11 changes the actual drive pattern after the drive pattern is learned by the learning unit 15, which will be explained later. The statement that the EMI noise is within the specified noise limit can also be rephrased as the EMI noise level being within the specified noise limit. The noise level can also be called the noise state.

[0030] Furthermore, the DP determination unit 11 is configured to be able to gradually change the actual drive pattern so that the learning unit 15 can learn from it. In other words, the DP determination unit 11 can set a learning drive pattern that has a gate current value different from the gate current value indicated by the actual drive pattern (drive pattern) associated with the operating conditions, using the actual drive pattern (drive pattern) as a reference. The drive pattern used when setting the learning drive pattern can also be called the reference pattern. The learning drive pattern corresponds to the learning drive mode.

[0031] Furthermore, the DP determination unit 11 can set multiple learning drive patterns for a single actual drive pattern. The gate current value indicated by each learning drive pattern will be a different value. Note that the DP determination unit 11 sets the learning drive pattern only when the learning unit 15 is learning. Also, the range of gate current values ​​to be different for each learning drive pattern is predetermined.

[0032] The noise and loss limits are determined by product specifications, etc. Furthermore, the EMI noise target value is a value that includes a margin (safety margin) above the noise limit.

[0033] EMI noise is noise generated in conjunction with the operation of the switching element 200. In this embodiment, radiated noise is used as an example of noise. The predicted EMI noise level is predicted by the EN prediction unit 12. The predicted EMI noise level is a predicted value of the noise level in EMI noise. The target EMI noise value is stored in the ENT storage unit 30. The target EMI noise value is a target value of the noise level in EMI noise.

[0034] In the following, the predicted EMI noise level will be referred to as the predicted value, and the target EMI noise level as the target value. The EMI noise level will also be referred to simply as the noise level. The EMI noise level corresponds to the noise generation state.

[0035] The EN prediction unit 12 obtains predicted values ​​from the monitor values ​​monitored by the SWF monitor unit 13, which will be described later. In this embodiment, the EN prediction unit 12 also uses EMI noise correlation data stored in the ENC storage unit 14, which will be described later, in addition to the monitor values ​​to obtain predicted values. The EN prediction unit 12 obtains predicted values ​​that correlate with the monitor values ​​in the EMI noise correlation data. The EN prediction unit 12 corresponds to the prediction unit. The EMI noise correlation data corresponds to the correlation data. Hereafter, the EMI noise correlation data will also be referred to as correlation data.

[0036] Furthermore, the EN prediction unit 12 predicts the EMI noise level, or it can be said that it predicts EMI noise. In addition, since the EN prediction unit 12 obtains predicted values ​​using the monitor values ​​monitored by the SWF monitor unit 13 and correlation data, it can be said that it can predict EMI noise in real time.

[0037] The SWF monitor unit 13 acquires monitor values ​​of the surge voltage waveform associated with the operation of the switching element 200. The SWF monitor unit 13 includes an AD converter. The SWF monitor unit 13 monitors the drain-source voltage (Vds) of the switching element 200 as the surge voltage waveform. The SWF monitor unit 13 also monitors, for example, the maximum value, minimum value, and slope of Vds.

[0038] Preferably, the SWF monitor unit 13 monitors the surge voltage waveform only during at least one of the turn-on and turn-off periods of the switching element 200. Alternatively, the SWF monitor unit 13 can be said to monitor the surge voltage waveform only during periods when Vds changes in accordance with the change in the drain-source voltage of the switching element 200. In other words, the SWF monitor unit 13 stops monitoring the surge voltage waveform outside of the turn-on and turn-off periods. This allows the drive IC 100 to reduce the current consumption of the SWF monitor unit 13.

[0039] The surge voltage waveform can also be described as data showing the surge voltage waveform. The SWF monitor unit 13 corresponds to the monitor unit. The SWF monitor unit 13 can also be described as the surge monitor unit.

[0040] The ENC memory unit 14 stores correlation data. The ENC memory unit 14 is part of the memory device. The correlation data shows the correlation between surge voltage waveforms and noise levels. The correlation data is pre-generated and stored in the ENC memory unit 14. The correlation data can be generated through experiments, etc. The ENC memory unit 14 corresponds to the correlation data storage unit.

[0041] Correlation data, for example, shows the correlation between surge voltage waveforms and noise levels at sampling points obtained by intermittently sampling the motor current (phase current) flowing through motor 300. Therefore, in correlation data, each of the multiple surge voltage waveforms is associated with a noise level correlated with that surge voltage waveform. In this way, the amount of data can be suppressed by generating correlation data through intermittent sampling.

[0042] Furthermore, the sampling points are preferably the maximum and minimum values ​​of the motor current. This helps to suppress the large amount of data in the correlation data. Also, correlation data can be generated at the location where EMI noise is maximum. The sampling points may also be near the maximum and minimum values ​​of the motor current.

[0043] Furthermore, it is preferable that the noise level be the noise level at a predetermined frequency. For example, the noise level is the noise level extracted only at the frequency that is a multiple (n times) of the frequency that determines the pulse width modulation period in the PWM drive of the inverter circuit. In other words, it is preferable that the correlation data shows the correlation between the noise level extracted only at the frequency that is a multiple of the carrier frequency for the inverter circuit and the surge voltage waveform. This helps to suppress the amount of data in the correlation data.

[0044] The learning unit 15 learns a drive pattern for each operating condition. Then, the learning unit 15 updates the pattern data (PD) stored in the PD memory unit 50 according to the learning results. More specifically, when the DP determination unit 11 changes the drive pattern to a learning drive pattern, the learning unit 15 updates the pattern data according to the difference between the predicted value and the target value when the current output circuit 20 is driven by the learning drive pattern. At this time, the learning unit 15 changes the learning drive pattern by associating it with the operating conditions associated with the actual drive pattern so that the noise level is within a specified range.

[0045] Furthermore, as described above, when the learning unit 15 is learning, the DP determination unit 11 sets up multiple learning drive patterns. The learning unit 15 evaluates the deviation between the predicted value and the target value while the switching element 200 is actually operating via the current output circuit 20 for each learning drive pattern. In other words, the learning unit 15 determines whether the noise level for each learning drive pattern is within a specified range. The learning unit 15 then determines that there is a learning effect if, as a result of repeating the driving and evaluation for each learning drive, the noise level is within a specified range and the slope of the surge voltage waveform monitored by the SWF monitor unit 13 is greater than that of the reference pattern, i.e., the switching loss is smaller than that of the reference pattern.

[0046] In this way, the learning unit 15 learns the optimal driving pattern for the operating conditions by repeatedly driving and evaluating with the learning driving pattern. Furthermore, it can be said that the learning unit 15 learns the driving pattern so that the optimal driving pattern for the operating conditions is associated even if there are product variations or aging degradation of the switching element 200 and the driving IC 100. It can also be said that the learning unit 15 updates the pattern data using supervised machine learning. In this case, the correspondence data between the learning driving pattern and the noise level obtained by repeatedly driving and evaluating can be considered as training data. The learning unit 15 may also update the pattern data as map data based on deviation and a specified range.

[0047] The computer 10 can also be called a processing unit or processor. The DP determination unit 11, EN prediction unit 12, SWF monitor unit 13, ENC memory unit 14, and learning unit 15 may be composed of the computer 10 and logic ICs.

[0048] The current output circuit 20 is a circuit that can adjust the switching speed of the switching element 200 by controlling the gate current output to the gate electrode. The current output circuit 20 is configured to adjust at least one of the turn-on switching speed and the turn-off switching speed. Multiple current sources are connected in parallel in the current output circuit 20 so that the gate current can be dynamically controlled. The current output circuit 20 is a circuit for performing active gate control.

[0049] The current output circuit 20 corresponds to the current control unit. The current output circuit 20 can also be called a drive output circuit, a speed change circuit, or a charging circuit. Furthermore, the current output circuit 20 can also be configured to dynamically control the gate current using a transistor and resistor instead of a current source.

[0050] The ENT memory unit 30 stores target values. These target values ​​are predetermined according to specifications and other factors. The ENT memory unit 30 can also be called a target memory unit. The ENT memory unit 30 is part of a memory device located outside the computer 10.

[0051] The OC monitor unit 40 monitors (acquires) sensor signals to determine the operating conditions of the switching element 200. The OC monitor unit 40 is electrically connected to the voltage sensor 501, the current sensor 502, and the temperature sensor 503. The OC monitor unit 40 acquires sensor signals indicating the battery voltage, sensor signals indicating the phase current, and sensor signals indicating the element temperature.

[0052] The OC monitor unit 40 acquires sensor signals for the corresponding phase. For example, if the drive IC 100 drives the U-phase switching element 100, it acquires a sensor signal indicating the phase current of the U-phase coil 301. Similarly, it acquires a sensor signal indicating the element temperature of the U-phase switching element 200. The OC monitor unit 40 can also be called a state monitor unit.

[0053] The PD memory unit 50 stores pattern data in which each of multiple operating conditions is associated with a drive pattern suitable for each operating condition. The pattern data can employ a lookup table or the like. Furthermore, the PD memory unit 50 is a storage medium whose contents can be updated. In other words, the PD memory unit 50 is configured to allow updating of the drive patterns associated with each operating condition. The PD memory unit 50 corresponds to a drive mode storage unit. The pattern data corresponds to drive mode data. The PD memory unit 50 is part of a storage device located outside the computer 10.

[0054] <Operation> The operation of the drive IC 100 will be explained using Figure 3. When the IG switch is turned ON (IGON), the drive IC 100 starts the process shown in the flowchart of Figure 3. The drive IC 100 then repeats the process from step S10 at a predetermined cycle until the IG switch is turned OFF (IGOFF).

[0055] In step S10, pattern data (PD) and operating conditions (OC) are acquired. In other words, the computer 10 receives pattern data from the PD storage unit 50 and operating conditions from the OC monitor unit 40. The pattern data and operating conditions are input to the DP determination unit 11. It can also be said that the DP determination unit 11 refers to the pattern data and operating conditions.

[0056] In step S11, a drive pattern is selected. The DP determination unit 11 selects a drive pattern associated with the operating conditions from among multiple drive patterns in the pattern data as the actual drive pattern.

[0057] In step S12, it is determined whether or not it is a learning session. If the DP determination unit 11 determines that it is a learning session, it proceeds to step S14; if it determines that it is not a learning session, it proceeds to step S13. The drive IC 100 performs learning processing at predetermined timings. Therefore, the DP determination unit 11 determines that it is a learning session when it is time to perform learning processing, and determines that it is not a learning session otherwise.

[0058] In step S13, the DP determination unit 11 drives the current output circuit 20. That is, if it is not a learning cycle, the DP determination unit 11 sets an actual drive pattern as the drive pattern for the current output circuit 20 according to the operating conditions and drives the current output circuit 20 with the actual drive pattern.

[0059] Step S14 is the same as step S10.

[0060] In step S15, the drive pattern is changed. The DP decision unit 11 sets a learning drive pattern for the learning unit 15 to learn from. The DP decision unit 11 sets the learning pattern based on the pattern data and operating conditions acquired in step S14. In other words, the DP decision unit 11 selects a drive pattern associated with the operating conditions acquired in step S14 as a reference pattern from among multiple drive patterns in the pattern data acquired in step S14. Then, the DP decision unit 11 sets the learning pattern from the reference pattern. Steps S14 to S24 are executed repeatedly as long as the learning cycle continues. In this case, the DP decision unit 11 sets a different learning pattern each time.

[0061] In step S16, the DP decision unit 11 drives the current output circuit 20. In other words, in the case of a learning cycle, the DP decision unit 11 drives the current output circuit 20 with a learning drive pattern.

[0062] In step S17, the SWF is monitored. The SWF monitoring unit 13 acquires the monitored value of the surge voltage waveform associated with the operation of the switching element 200. The SWF monitoring unit 13 outputs the acquired monitored value to the EN prediction unit 12. It can also be said that the SWF monitoring unit 13 makes the acquired monitored value available for reference by the EN prediction unit 12.

[0063] In step S18, the ENC is obtained. The EN prediction unit 12 obtains correlation data (ENC) from the ENC storage unit 14. It can also be said that the EN prediction unit 12 refers to the correlation data stored in the ENC storage unit 14.

[0064] In step S19, the EN is predicted. The EN prediction unit 12 predicts the EMI noise (EN). That is, the EN prediction unit 12 obtains a predicted value using the monitor value and correlation data.

[0065] In step S20, ENT is obtained. The computer 10 obtains the EMI noise target value (ENT) from the ENT memory unit 30.

[0066] In step S21, feedback is provided. The computer 10 feeds back the predicted value from the EN prediction unit 12 and the target value from the ENT memory unit 30 to the DP determination unit 11. It can also be said that the computer 10 inputs the discrepancy between the predicted value and the target value when driven by the learning drive pattern set in step S15 to the DP determination unit 11.

[0067] In step S22, learning is performed for each operating condition. The learning unit 15 executes the learning process for each operating condition.

[0068] In step S23, it is determined whether or not there is a learning effect. The learning unit 15 determines whether or not there is a learning effect by comparing the deviation with a specified range. After repeating driving and evaluation with each learning drive pattern, the learning unit 15 determines that there is a learning effect if there is a learning drive pattern in which the noise level is within the specified range and the slope of the surge voltage waveform monitored by the SWF monitor unit 13 is greater than that of the reference pattern, i.e., the switching loss is smaller than that of the reference pattern, and proceeds to step S24. If the deviation is not within the specified range, the learning unit 15 determines that there is no learning effect and returns to step S10.

[0069] In step S24, the PD is updated. The learning unit 15 updates the pattern data in the PD memory unit 50.

[0070] <Effects> As described above, the drive IC 100 obtains a predicted value from the monitored value of the surge voltage waveform associated with the operation of the switching element 200. Then, the drive IC 100 changes the actual drive pattern according to the operating conditions so that the noise and loss are within the specified limits, depending on the deviation between the predicted value and the target value. In this way, the drive IC 100 can achieve both noise suppression and switching loss suppression. Furthermore, the power converter 1 includes the drive IC 100 and the switching element 200 driven by the drive IC 100. In this way, the power converter 1 can achieve both noise suppression and switching loss suppression.

[0071] Furthermore, the drive IC 100 updates its pattern data based on the deviation and specified range when the actual drive pattern is changed to a learning drive pattern by the learning unit 15, as described above. Therefore, even if there are product variations or deterioration over time in the switching element 200 and the drive IC 100, the drive IC 100 can set the optimal drive pattern for the operating conditions.

[0072] Preferred embodiments of the present disclosure have been described above. However, the present disclosure is not limited to the above embodiments, and various modifications are possible without departing from the spirit of the present disclosure. A second embodiment, as another form of the present disclosure, is described below. The above embodiments and the second embodiment can be implemented individually, or they can be implemented in combination as appropriate. The present disclosure can be implemented in various combinations, not limited to the combinations shown in the embodiments.

[0073] (Second Embodiment) The drive IC 101 of the second embodiment will be described using Figures 4 and 5. Here, the differences between the drive IC 101 and the drive IC 100 will be mainly explained. The drive IC 101 differs from the drive IC 100 in that it predicts the motor surge voltage applied to the motor 300. In Figure 4, the same components as in Figure 2 are given the same reference numerals. In Figure 5, the same processes as in Figure 3 are given the same step numbers. The power converter 1 of the second embodiment is equipped with the drive IC 101.

[0074] In this embodiment, the motor surge voltage, which is generated by the operation of the switching element 200 and applied to the motor 300, corresponds to noise. Also in this embodiment, the motor surge voltage corresponds to the noise level. Hereinafter, the motor surge voltage will also be simply referred to as motor surge. MS is an abbreviation for motor surge. MTF is an abbreviation for motor transfer function. MST is an abbreviation for motor surge target value. Note that the motor transfer function corresponds to the transfer function.

[0075] Computer 10a includes an MS prediction unit 12a and an MTF storage unit 14a. The MS prediction unit 12a is provided in place of the EN prediction unit 12. The MS prediction unit 12a obtains a predicted value of the motor surge voltage from the monitored value of the surge voltage waveform. In this embodiment, in addition to the monitored value, the MS prediction unit 12a is also used to obtain the predicted value using the motor transfer function stored in the MTF storage unit 14a. The MS prediction unit 12a calculates the motor surge voltage correlated with the monitored value from the motor transfer function and the monitored value. The MS prediction unit 12a obtains the motor surge voltage as a predicted value through this calculation. The MS prediction unit 12a corresponds to the prediction unit. Hereafter, the predicted value of the motor surge voltage will also be simply referred to as the predicted value.

[0076] Furthermore, the MS prediction unit 12a can be said to predict motor surge voltage, or even predict motor surges. In addition, since the MS prediction unit 12a obtains predicted values ​​using the monitored values ​​of the surge voltage waveform monitored by the SWF monitor unit 13 and the motor transfer function, it can be said that it can predict motor surges in real time.

[0077] The MTF memory unit 14a is provided in place of the ENC memory unit 14. The MTF memory unit 14a stores the motor transfer function of the motor 300, which has been generated in advance. The MTF memory unit 14a corresponds to the transfer function memory unit.

[0078] The drive IC 101 is equipped with an MST memory unit 30a. The MST memory unit 30a is provided in place of the ENT memory unit 30. The MST memory unit 30a stores a target value. The target value is predetermined according to specifications, etc. The MST memory unit 30a stores the motor surge target value as the target value. The MST memory unit 30a can also be called the target acquisition unit. The motor surge target value is the target value of the motor surge voltage. Hereafter, the motor surge target value will also be referred to as the target value.

[0079] Therefore, the DP determination unit 11 changes the drive pattern according to the deviation between the predicted value and the target value of the motor surge. In other words, the DP determination unit 11 changes the actual drive pattern according to the deviation so that the motor surge is within a specified range and the switching loss of the switching element 200 is within a specified loss value. The statement that the motor surge is within a specified value can also be rephrased as the motor surge being within a specified value of the motor surge voltage.

[0080] The operation of the drive IC 101 will be explained using Figure 5. The drive IC 101 executes step S18a after step S17.

[0081] In step S18a, the MTF is obtained. The MS prediction unit 12a obtains the motor transfer function (MTF) from the MTF storage unit 14a.

[0082] In step S19a, the MS is predicted. The MS prediction unit 12a predicts the motor surge (MS). That is, the MS prediction unit 12a obtains a predicted value using the monitored value and the motor transfer function.

[0083] In step S20a, the MST is obtained. The computer 10 obtains the motor surge target value (MST) from the MST storage unit 30a.

[0084] In step S21a, feedback is provided. The computer 10 feeds back the predicted value from the MS prediction unit 12a and the target value from the MST memory unit 30a to the DP determination unit 11. It can also be said that the computer 10 inputs the discrepancy between the predicted value and the target value when driven by the learning drive pattern set in step S15 to the DP determination unit 11.

[0085] The drive IC 101 can achieve the same effect as the drive IC 100. Furthermore, if the motor surge increases, there is a possibility of malfunction in the motor 300. However, the drive IC 101 can suppress malfunctions in the motor 300. Malfunctions in the motor 300 include, for example, damage to the insulation of the copper wires in the motor 300 or the copper wires connecting the motor 300 to the power converter 1.

[0086] Furthermore, the drive IC 100 and drive IC 101 can also be used in combination. In this case, the drive IC comprises the components of drive IC 100 and drive IC 101. Specifically, the drive IC includes an EN prediction unit 12, an ENC memory unit 14, an ENT memory unit 30, an MS prediction unit 12a, an MTF memory unit 14a, and an MST memory unit 30a. The drive IC changes the drive pattern according to the deviation between the predicted value and the target value in EMI noise, and the deviation between the predicted value and the target value in motor surge.

[0087] This disclosure is described in accordance with embodiments, but it is understood that this disclosure is not limited to such embodiments or structures. This disclosure also includes various modifications and variations within the scope of equivalents. In addition, while various combinations and forms are shown in this disclosure, other combinations and forms that include one, more, or fewer of those elements also fall within the scope and idea of ​​this disclosure. [Explanation of symbols]

[0088] 1...Power converter, 10,10a...Computer, 11...DP determination unit, 12...EN prediction unit, 12a...MS prediction unit, 13...SWF monitor unit, 14...ENC memory unit, 14a...MTF memory unit, 15...Learning unit, 20...Current output circuit, 30...ENT memory unit, 30a...MST memory unit, 40...OC monitor unit, 50...PD memory unit, 100,101...Drive IC, 200...Switching element, 300...Motor, 301...U-phase coil, 302...V-phase coil, 303...W-phase coil, 400...Battery, 501...Voltage sensor, 502...Current sensor, 503...Temperature sensor

Claims

1. A drive circuit for driving a switching element (200) having a gate electrode, A current control unit (20) that can adjust the switching speed by controlling the gate current output to the gate electrode, A drive mode setting unit (11) sets the actual drive mode of the current control unit according to the operating conditions of the switching element, and drives the current control unit in the actual drive mode, A monitoring unit (13) acquires a monitor value of the surge voltage waveform associated with the operation of the switching element, The system includes a prediction unit (12, 12a) that acquires a predicted value of the noise level, which is the state of noise generated in conjunction with the operation of the switching element, from the monitor value, The drive mode setting unit changes the actual drive mode according to the operating conditions, in accordance with the deviation between the predicted value and the target value of the noise level, so that the noise is within a specified range and the switching loss of the switching element is within a specified loss value. Furthermore, a drive mode storage unit (50) stores drive mode data in which each of the multiple operating conditions is associated with the drive mode suitable for each operating condition, A drive circuit comprising: when the drive mode setting unit changes the drive mode to a learning drive mode different from the actual drive mode, a learning unit (15) updates the drive mode data by associating the learning drive mode with the operating conditions associated with the actual drive mode when the deviation between the predicted value and the target value in the state in which the current control unit is driven in the learning drive mode falls within the specified range.

2. The noise is radiated noise generated in conjunction with the operation of the switching element. The drive circuit according to claim 1, wherein the noise level is the noise level of the radiated noise.

3. The system further includes a correlation data storage unit (14) in which correlation data showing the correlation between the surge voltage waveform and the noise level is pre-generated and stored. The drive circuit according to claim 2, wherein the prediction unit acquires the noise level correlated with the monitor value in the correlation data as the predicted value.

4. The aforementioned switching element is included in an inverter circuit that drives a motor. The drive circuit according to claim 3, wherein the correlation data shows the correlation between the surge voltage waveform and the noise level at sampling points obtained by intermittently sampling the motor current flowing through the motor.

5. The drive circuit according to claim 4, wherein the sampling points are the maximum and minimum values ​​of the motor current.

6. The drive circuit according to claim 4, wherein the correlation data shows the correlation between the noise level extracted only at the carrier frequency multiplication frequency for the inverter circuit and the surge voltage waveform.

7. The aforementioned switching element is included in an inverter circuit that drives a motor. The noise is a motor surge voltage applied to the motor that is generated in conjunction with the operation of the switching element. The drive circuit according to claim 1, wherein the noise level is the motor surge voltage.

8. The system further comprises a transfer function storage unit (14a) in which the transfer function of the motor is pre-generated and stored, The drive circuit according to claim 7, wherein the prediction unit obtains the motor surge voltage as the predicted value by calculating the motor surge voltage correlated with the monitor value from the transfer function and the monitor value.

9. The drive circuit according to claim 1, wherein the monitoring unit stops monitoring the surge voltage waveform except during the turn-on and turn-off periods of the switching element, and operates only during at least one of the turn-on and turn-off periods of the switching element to acquire the monitored value.

10. A power conversion device comprising a drive circuit as described in claim 1, and a switching element driven by the drive circuit.

Images