Inverter control device and power conversion device
By calculating an estimated DC current using AC sensor values and duty cycle information, the system addresses the challenge of reliable DC sensor diagnosis in power conversion devices, ensuring continuous and accurate fault detection.
Patent Information
- Application Number
- DE112017001441
- Authority / Receiving Office
- DE · DE
- Patent Type
- Patents
- Current Assignee / Owner
- Priority Date
- 2016-05-17
- Filing Date
- 2017-04-11
- Publication Date
- 2025-12-04
- Estimated Expiration
- 2037-04-11
AI Technical Summary
Existing systems face challenges in reliably diagnosing current sensors in power conversion devices for hybrid and electric vehicles due to time delays between direct and alternating current changes, leading to potential false anomaly detections.
The system calculates an estimated DC current value using AC sensor values and duty cycle information to continuously diagnose the DC sensor, minimizing time delays and ensuring accurate fault detection.
This approach allows for continuous and reliable diagnosis of DC sensors, reducing the risk of false anomaly detection and maintaining high diagnostic reliability.
Smart Images

Figure 00000000_0000_ABST
Abstract
Description
Technical field
[0001] The present invention relates to an inverter control device and a power conversion device, and in particular to an inverter control device and a power conversion device for driving a vehicle. Technical background
[0002] A power conversion device is used to drive a motor in hybrid and electric vehicles. The power conversion device uses a power semiconductor in an inverter circuit to convert direct current supplied by a battery into alternating current to drive the motor.
[0003] The power conversion device includes a DC current sensor, which measures the DC current supplied to the inverter circuit by the battery, in addition to a three-phase AC current sensor, which measures a current flowing in the motor.
[0004] In recent years, the need to detect anomalies / faults in the power conversion device has increased due to the implementation of functional safety standards for motor vehicles. Therefore, it is necessary to perform diagnostic processing for the DC sensor to detect anomalies / faults.
[0005] There is PTL 1 as a technical background in the present technical field. PTL 1 discloses that "when a current command value for a PDU 3 is specified as a torque command value of a motor 1, the feedback computer devices 14a and 14b compare a phase current to be supplied to the motor 1 with the current command value, calculating a voltage command value in such a way that the motor 1 is driven according to the torque command value. A voltage loss correction device of a current sensor disturbance detector 19 corrects a loss in the PDU 3 with respect to the voltage command value and calculates a voltage command value after correction that is actually supplied to the motor 1."Because an inverter current estimator of the current sensor fault detector 19 calculates a PDU current estimated to flow to the PDU 13 based on the current command value and the voltage command value after correction, the current sensor determiners of the current sensor fault detector 19 compare a battery current detected by a battery current sensor 5 and a PDU current to detect the presence or absence of an anomaly in the battery current sensor 5.
[0006] PTL 2 discloses a generic inverter control device. PTL 3 discloses a power steering system with a brushless three-phase motor, in which an abnormality of a control unit is detected based on a DC busbar current of an inverter, determined by a current sensor, and an estimated main current. State of the art patent literature PTL 1: JP 3 795 447 B PTL 2: JP 5 760 778 B2 PTL 3: DE 10 2013 217 970 A1 Summary of the invention; Technical task
[0007] In the PTL 1, the PDU current is calculated using the current command value and the voltage command value after loss correction. The PDU current is then compared to the battery current measured by the battery current sensor, thus diagnosing any anomalies in the battery current sensor. However, a certain amount of time is required for the actual battery current to change after the current command value or the voltage command value has been modified.If the diagnostic procedure is performed according to the present system immediately after the current command value or the voltage command value has been changed, or immediately after the target torque, which is a factor in determining the current command value or the voltage command value, has been changed, a discrepancy will occur between a PDU current value and a battery current value, even if the battery current sensor is functioning normally. This carries the risk of a false battery current sensor anomaly detection. Therefore, the present system has a problem in that the battery current sensor diagnostic procedure cannot be performed immediately after a change in the target torque.
[0008] It is an object of the present invention to perform a diagnosis of a current sensor while maintaining high reliability. Solution to the task
[0009] According to a first aspect of the invention, an inverter control device according to claim 1 is provided.
[0010] According to a second aspect of the invention, a power conversion device according to claim 5 is provided. Further aspects of the invention are the subject of the dependent claims, the drawings, and the description of exemplary embodiments. The advantageous effects of the invention
[0011] According to the present invention, the diagnosis of a current sensor can be performed while maintaining high reliability. Brief description of the drawings Fig. Figure 1 is a graphical representation illustrating a configuration example of a power conversion device and a peripheral circuit in a first embodiment. Fig. Figure 2 is a graphical representation illustrating a configuration example of an inverter circuit in the first embodiment. Fig. Figure 3 is a graphical representation illustrating a flowchart of diagnostic processing of a DC sensor in the first embodiment. Fig. Figure 4 is a graphical representation illustrating a configuration example of a power conversion device and a peripheral circuit in a second embodiment. Fig. Figure 5 is a graphical representation illustrating a configuration example of an inverter circuit in the second embodiment. Fig. Figure 6 is a graphical representation illustrating a flowchart of diagnostic processing of a DC sensor in the second embodiment. Fig. Figure 7 is a graphical representation illustrating a configuration example of a power conversion device and a peripheral circuit in a third embodiment. Fig. Figure 8 is a graphical representation illustrating a flowchart of diagnostic processing of a DC sensor in the third embodiment. Fig. Figure 9 is a graphical representation illustrating an example of a correction effect in a state where an AC sensor value lus of the U-phase has an offset error loff. Fig. Figure 10 is a graphical representation showing a difference in the correction effect according to the values of the correction coefficients K in the example below. Fig. 9 illustrates. Fig. Figure 11 is a graphical representation illustrating an example of a correction effect in a state where an AC sensor value lus of the U-phase and an AC sensor value Ivs of the V-phase have an offset error loff. Fig. Figure 12 is a graphical representation showing a difference in the correction effect according to the values of the correction coefficients K in the example below. Fig. 11 illustrated. Fig. Figure 13 is a graphical representation illustrating an example of a correction effect in a state where an AC sensor value lus of the U-phase and an AC sensor value Ivs of the V-phase have an offset error loff. Fig. Figure 14 is a graphical representation showing the difference in the correction effect according to the values of the correction coefficients K in the example below. Fig. 13 illustrated. Fig. Figure 15 is a graphical representation illustrating an example of a correction effect in a state where an AC sensor value lus of the U-phase has a gain error Ig. Fig. Figure 16 is a graphical representation showing a difference in the correction effect according to the values of the correction coefficients K in the example below. Fig. 15 illustrated. Description of the embodiments
[0012] In the present embodiment, an estimated DC current value is calculated from an AC sensor value and the duty cycle values of the phases, and a diagnosis of a DC sensor is performed based on a DC sensor value output by a DC sensor and an estimated DC current value.
[0013] Because the time delay between the direct current flowing from the battery to the inverter circuit and the alternating current flowing in the motor is small, the deviation between the estimated direct current value and the DC sensor reading is small, unless the DC sensor is faulty. Therefore, the present method allows for continuous diagnosis of the DC sensor.
[0014] The problems, configurations, and effects, with the exception of those described above, are clarified by the description of the following embodiments. The embodiments are described below with respect to the drawings. The first embodiment
[0015] The present embodiment describes an example of a power conversion device that can continuously diagnose the anomaly of a DC sensor.
[0016] Fig. Figure 1 is a graphical representation illustrating a configuration example of a power conversion device and a peripheral circuit in a first embodiment.
[0017] A DC power source 3 is a power source for driving a motor 2 and corresponds, for example, to a battery. A power conversion device 1 converts the DC power received from the DC power source 3 into AC power to drive the motor 2. Furthermore, the power conversion device 1 also has a function to convert the power of the motor 2 into DC power to charge the DC power source 3.
[0018] Motor 2 is a three-phase electric motor with three windings. Furthermore, an angle sensor (not illustrated) for measuring the motor's rotation angle is installed in motor 2, the angle sensor outputting a measured rotation angle of motor 2 as an angle sensor value 7 to the power conversion device 1.
[0019] The power conversion device 1 includes an inverter control device 16, an inverter circuit 9, the AC sensors 14a to 14c, a voltage sensor 10 and a DC sensor 12.
[0020] As in Fig. As illustrated in Figure 2, the inverter circuit 9 contains two power semiconductors, one for each winding (phase) of the motor 2. The power semiconductor corresponds, for example, to a power metal-oxide-semiconductor field-effect transistor (power MOSFET), an insulated-gate bipolar transistor (IGBT), or the like. Furthermore, the inverter circuit 9 includes a drive circuit for switching the power semiconductor ON / OFF for each power semiconductor. Because the motor 2 has three phases, the inverter circuit 9 has a total of six power semiconductors 90a to 90f and six drive circuits 91a to 91f. It is stated that in the present embodiment, the upper power semiconductors 90a, 90c, and 90e are collectively referred to as an upper arm, while the lower power semiconductors 90b, 90d, and 90f are collectively referred to as a lower arm. Additionally, the inverter circuit 9 contains a smoothing capacitor 92.
[0021] The control circuits 91a to 91f switch the power semiconductors 90a to 90f ON / OFF on the basis of the pulse width modulation signals (PWM signals) 22a to 22f, which are output by the inverter control device 16.
[0022] The smoothing capacitor 92 is a capacitor for smoothing the current generated by the ON / OFF of the power semiconductor and for suppressing the ripple of the DC current supplied by the DC power source 3 of the inverter circuit 9 and corresponds, for example, to an electrolytic capacitor or a film capacitor.
[0023] The AC sensors 14a to 14c are sensors for measuring the alternating currents flowing in the phases (a U-phase, a V-phase, a W-phase) of the motor 2. The AC sensor 14a measures an alternating current lu flowing in the U-phase and outputs an AC sensor value lus to the inverter control circuit 16. Similarly, the AC sensor 14b measures an alternating current Iv flowing in the V-phase and outputs an AC sensor value Ivs to the inverter control device 16. The AC sensor 14c measures an alternating current Iw flowing in the W-phase and outputs an AC sensor value Iws to the inverter control device 16.
[0024] It is stated that in the present embodiment, the AC current sensors 14a to 14c are installed to measure the current flowing from the inverter circuit 9 into the motor 2 as a positive current value. However, the AC current sensors 14a to 14c can also be installed to measure a current value opposite to the positive current value.
[0025] The voltage sensor 10 is a sensor that measures an output voltage of the DC power source 3 and outputs a measured voltage value as a voltage sensor value 11 to the inverter control device 16.
[0026] The DC current sensor 12 is a sensor that measures a DC current Idc flowing between the DC power source 3 and the inverter circuit 9, and outputs a measured current value as a DC current sensor value Idcs to the inverter control device 16. It is stated that in the present embodiment, the DC current sensor 12 is installed to measure the current flowing from the DC power source 3 into the inverter circuit 9 as a positive current value. However, the DC current sensor 12 can be installed to measure a current value opposite to the positive current value.
[0027] The inverter control device 16 communicates with an electronic control device (not illustrated) outside the power conversion device 1 and receives a target torque 5 of the motor 2 from another electronic control device. The inverter control device 16 switches the PWM signals 22a to 22f to control the inverter circuit 9 to drive the motor 2 based on the target torque 5 and the AC sensor values lus, Ivs, and Iws. If the inverter control device 16 determines that a fault has occurred within the power conversion device 1, the inverter control device 16 further outputs an anomaly message signal 6 to an external anomaly message device 4.
[0028] The inverter control device 16 internally contains a (not illustrated) communication circuit, a target current calculation unit 17, a duty cycle calculation unit 19, a PWM signal generation unit 21, a motor speed calculation unit 23 and a DC current sensor diagnostic unit 25.
[0029] The target current calculation unit 17 calculates a current value flowing into the motor 2, using the target torque 5, the voltage sensor value 11, and a motor speed value 24 output by the motor speed calculation unit 23, and outputs the current value as a target current value 18 to the duty cycle calculation unit 19. The target current value 19 contains information on a d-axis target current value and a q-axis target current value.
[0030] The duty cycle calculation unit 19 calculates a U-phase duty cycle value Du, a V-phase duty cycle value Dv, and a W-phase duty cycle value Dw based on the target current value 18 output by the target current calculation unit 17 and the AC sensor values lus, Ivs, and Iws, and outputs the U-phase duty cycle value Du, the V-phase duty cycle value Dv, and the W-phase duty cycle value Dw to the PWM signal generation unit 21 and the DC sensor diagnostic unit 25. Du represents an ON-time ratio of the power semiconductor 90a, while an ON-time ratio of the power semiconductor 90b, which is paired with the power semiconductor 90a, is represented by 1 - Du. Similarly, Dv represents an ON-time ratio of the 90c power semiconductor, while an ON-time ratio of the 90d power semiconductor is represented by 1- Dv.Similarly, Dw represents an ON-time ratio of the 90e power semiconductor, while an ON-time ratio of the 90f power semiconductor is represented by 1 - Dw.
[0031] The PWM signal generation unit 21 contains a (not illustrated) timer and generates the PWM signals 22a to 22f based on a timer value and the duty cycle values Du, Dv and Dw and outputs the PWM signals 22a to 22f to the inverter circuit 9.
[0032] Furthermore, the PWM signal generation unit 21 controls the PWM signals 22a to 22f in such a way that the motor 2 is not driven when the anomaly message signal 6 is output by the DC sensor diagnostic unit 25. An example of a state in which the motor 2 is not driven is a state in which all six power semiconductors in the inverter circuit 9 are switched OFF. Other examples include a state of the six power semiconductors in which the three power semiconductors in the upper arm are switched ON and the three power semiconductors in the lower arm are switched OFF, or conversely, a case in which the three power semiconductors in the upper arm are switched OFF and the three power semiconductors in the lower arm are switched ON.
[0033] The motor speed calculation unit 23 calculates a motor speed from a change in the angle sensor value 7 and outputs a calculated motor speed value 24 to the target current calculation unit 17.
[0034] The DC sensor diagnostic unit 25 is a unit that performs a fault diagnosis of the DC sensor 12 and internally contains a calculation unit 26 for the estimated DC current and a comparison unit 28. The calculation unit 26 for the estimated DC current calculates an estimated DC current value Idce1 based on expression (1) using the duty cycle values Du, Dv, and Dw and the AC sensor values lus, Ivs, and Iws of the phases. The calculation unit 26 for the estimated DC current outputs the calculated estimated DC current value Idce1 to the comparison unit 28. [Expression 1] Idce1=Du×Ius+Dv×Ivs+Dw×Iws
[0035] Here, expression (1) is obtained according to the following idea. In Fig. 2. The direct current Idc flowing into the direct current sensor 12 is an average value of the currents Idca between the smoothing capacitor 92 and the power semiconductors. This current Idca is a sum of the currents lu1, Iv1, and Iw1 flowing in the power semiconductors 90a, 90c, and 90e, according to Kirchhoff's law. When the power semiconductors 90a, 90c, and 90e are ON, the currents lu1, Iv1, and Iw1 become equal to the alternating current values lu, Iv, and Iw, respectively. When the power semiconductors 90a, 90c, and 90e are OFF, the currents lu1, Iv1, and Iw1 are 0. Therefore, the average values of the currents lu1, Iv1, and Iw1 are obtained by multiplying the AC values lu, Iv, and Iw by the ON-time ratios of the power semiconductors 90a, 90c, and 90e. Because the ON-time ratios of the power semiconductors 90a, 90c, and 90e correspond to the duty cycle values Du, Dv, and Iw1, respectively, the average values of the currents lu1, Iv1, and Iw1 are obtained by multiplying the AC values lu, Iv, and Iw by the ON-time ratios of the power semiconductors 90a, 90c, and 90e.If Dw are and the AC current values lu, Iv and Iw can be replaced by the AC current sensor values lus, Ivs and Iws, the DC current Idc can be estimated by calculating expression (1).
[0036] The comparator unit 28 compares the estimated DC current value Idce1, which is output by the computation unit 26 for the estimated DC current, and the DC current sensor value Idcs and determines, based on a comparison result, whether the DC current sensor 12 is anomalous. In the case where the comparator unit 28 determines that the DC current sensor 12 is anomalous, the comparator unit 28 outputs the anomaly message signal 6 to the anomaly message device 4.
[0037] It is stated that in the present embodiment, the DC sensor diagnostic unit 25 is illustrated within the inverter control device 16. However, the DC sensor diagnostic unit 25 can be provided outside the inverter control device 16. Furthermore, the DC sensor diagnostic unit 25 can be mounted in an electronic control device separate from the power conversion device 1.
[0038] The anomaly notification device 4 receives the anomaly notification signal 6 from the inverter control device 16 and notifies an occupant of the occurrence of the anomaly. Examples of an anomaly notification methods include a method of illuminating a lamp, a method of generating an alarm tone, and a method of giving a voice message.
[0039] Fig. Figure 3 is a flowchart illustrating the diagnostic processing of the DC sensor 12 in the first embodiment. This diagnostic processing is performed by the inverter control device 16 with an arbitrary time control.
[0040] In step S100, the inverter control device 16 acquires the AC sensor values lus, Ivs and Iws and the DC sensor value Idcs. Then, in step S101, the computation unit 26 calculates the estimated DC current value Idce1 based on expression (1) using the duty cycle values Du, Dv and Dw and the AC sensor values lus, Ivs and Iws.
[0041] Next, in step S102, the comparator unit 28 compares the DC current sensor value Idcs and the estimated DC current value Idce1, determining whether the difference between the DC current sensor value Idcs and the estimated DC current value Idce1 is a threshold value of 1 or greater. If the difference is a threshold value of 1 or greater, the comparator unit 28 determines that the DC current sensor 12 is anomalous and proceeds to processing in step S103. If the difference is less than a threshold value of 1, the inverter control device 16 terminates the diagnostic processing.
[0042] In step S103, the comparator unit 28 outputs the anomaly message signal 6 to the PWM signal generation unit 21 and the anomaly message device 4. Then, in step S104, the PWM signal generation unit 21 processes the switching of the PWM signals 22a to 22f in such a way that the motor 22 is not driven. Afterwards, the inverter control device 16 terminates the diagnostic processing.
[0043] It is stated that the output of the anomaly message signal 6 to the PWM signal generation unit 21 in step S103 and the control of the PWM signals in step S104 are not essential. After the determination that the DC sensor 12 is anomalous has been carried out, only the output of the anomaly message signal 6 to the anomaly message device 4 can be performed.
[0044] Furthermore, during diagnostic processing, after Fig. 3. The determination that the DC current sensor 12 is anomalous is made if the difference between the DC current sensor value Idcs and the estimated DC current value Idce1 is a threshold of 1 or greater. However, whether the DC current sensor 12 is anomalous can also be determined using a further criterion. For example, if the difference between the DC current sensor value Idcs and the estimated DC current value Idce1 is greater than a difference from the previous diagnosis by a fixed value or more, the determination that the DC current sensor 12 is anomalous can be made.
[0045] According to the present embodiment, the calculation unit 26 calculates the estimated DC current value Idce1 based on the AC sensor values lus, Ivs, and Iws and the duty cycle values Du, Dv, and Dw. The comparison unit 28 then compares the estimated DC current value Idce1 with the DC sensor value Idcs, thereby performing a diagnostic check of the DC sensor. Because both the AC sensor values lus, Ivs, and Iws and the DC sensor value Idcs change in connection with the operation of the inverter circuit 9, the time difference between the sensors is small. Therefore, the difference between the estimated DC current Idce1 and the DC sensor value Idcs is small even immediately after a change in the target torque, thus ensuring continuous diagnostic check of the DC sensor 12. The second embodiment
[0046] In the present embodiment, an example of a power conversion device is shown which can continuously diagnose the anomaly of a DC sensor even in a motor and in an inverter circuit which have different configurations from the first embodiment.
[0047] Fig. Figure 4 illustrates a configuration example of the power conversion device and a peripheral circuit in the second embodiment. It is stated that the same elements as in the configuration example in the first embodiment are given the same reference numerals, with the description of these elements omitted.
[0048] In the second embodiment, a motor 2a is a three-phase, six-wire electric motor, different from the motor 2 in the first embodiment. Furthermore, in the second embodiment, a power conversion device 1a includes an inverter circuit 9a, different from that of the first embodiment, and an inverter control device 16a, also different from that of the first embodiment. The number of wires in the inverter circuit 9a and the motor 2a is two for each of the U-phase, V-phase, and W-phases, for a total of six. Each of the AC current sensors 14a to 14c is installed for one wire of each phase. This is because it is not necessary to measure the current values of both wires of each phase, as the currents in the two wires of each phase are of the same magnitude but flow in opposite directions.For this reason, no AC current sensors are installed in three of the wires of the inverter circuit 9a and the motor 2a. Additionally, the PWM signals output by the inverter control device 16a to the inverter circuit 9a are increased to twelve lines.
[0049] Fig. Figure 5 is a graphical representation illustrating a configuration example of the inverter circuit 9a in the second embodiment. It is stated that the same elements as in the configuration example in the first embodiment are given the same reference numerals, with the description of these elements omitted.
[0050] Because the motor 2a is a three-phase six-wire electric motor, the inverter circuit 9a internally contains twelve power semiconductors 90a to 90l and twelve control circuits 91a to 91l. The control circuits 91a to 91l switch the power semiconductors 90a to 90l ON / OFF based on the PWM signals 22a to 22l output by the inverter control device 16a.
[0051] The inverter control device 16a of the second embodiment includes a target current calculation unit 17a, a duty cycle calculation unit 19a, a PWM signal generation unit 21a and a DC sensor diagnostic unit 25a, which differ from those of the first embodiment.
[0052] The target current calculation unit 17a calculates a current value flowing into the motor 2a using the target torque 5, a voltage sensor value 11, and a motor speed value 24 output by a motor speed calculation unit 23, and outputs the current value to the duty cycle calculation unit 19a as a target current value 18a. The target current value 18a contains information on a d-axis target current value, a q-axis target current value, and a 0-axis target current value. The 0-axis current refers to a sum of the alternating currents of the U-phase, the V-phase, and the W-phase.
[0053] The duty cycle calculation unit 19a calculates the timing information Cu, Cv, and Cw for switching the PWM signals of the U-phase, V-phase, and W-phase based on the target current value 18a output by the target current calculation unit 17a and the AC sensor values lus, Ivs, and Iws. It then outputs the timing information Cu, Cv, and Cw for switching the PWM signals to the PWM signal generation unit 21a. In the case of a three-phase, six-wire system, the rise and fall of the PWM signal can occur multiple times within a PWM cycle, in which case the timing information is output instead of the duty cycle values.
[0054] Furthermore, the duty cycle calculation unit 19a calculates the duty cycle values Du1 and Du2 of the U-phase, the duty cycle values Dv1 and Dv2 of the V-phase, and the duty cycle values Dw1 and Dw2 of the W-phase and outputs the duty cycle values Du1 and Du2 of the U-phase, the duty cycle values Dv1 and Dv2 of the V-phase, and the duty cycle values Dw1 and Dw2 of the W-phase to the DC current sensor diagnostic unit 25a. Du1 represents an ON-time ratio of a power semiconductor 90a, while an ON-time ratio of a power semiconductor 90b, which is arranged in pairs with the power semiconductor 90a, is represented by 1-Du1. Furthermore, Du2 represents an ON-time ratio of a power semiconductor 90g, while an ON-time ratio of a power semiconductor 90h, which is arranged in pairs with the power semiconductor 90g, is represented by 1 - Du2.Similarly, Dv1 represents an ON-time ratio of a 90c power semiconductor, while Dv2 represents an ON-time ratio of a 90i power semiconductor. Dw1 represents an ON-time ratio of a 90e power semiconductor, while Dw2 represents an ON-time ratio of a 90k power semiconductor.
[0055] The PWM signal generation unit 21a generates the PWM signals 22a to 22l based on a value of an internal timer and the time control information Cu, Cv and Cw of switching the PWM signals output by the duty cycle calculation unit 19a and outputs the PWM signals 22a to 22l to the inverter circuit 9a.
[0056] Furthermore, the PWM signal generation unit 21a controls the PWM signals 22a to 22l in such a way that the motor 2a is not driven if an anomaly message signal 6 is output by the DC sensor diagnostic unit 25a. An example of a state in which the motor 2a is not driven is a state in which all twelve power semiconductors in the inverter circuit 9a are switched OFF. Other examples include a state of the twelve power semiconductors in which the six power semiconductors in an upper arm are switched ON and the six power semiconductors in a lower arm are switched OFF, or conversely, a case in which the six power semiconductors in the upper arm are switched OFF and the six power semiconductors in the lower arm are switched ON.
[0057] The DC sensor diagnostic unit 25a includes a calculation unit 26a for the estimated DC value and a comparison unit 28, which differ from the first embodiment.
[0058] The calculation unit 26a for the estimated DC value calculates an estimated DC value Idce2 based on expression (2) using the duty cycle values Du1, Du2, Dv1, Dv2, Dw1 and Dw2 calculated by the duty cycle calculation unit 19a and the AC sensor values lus, Ivs and Iws. The calculation unit 26a for the estimated DC outputs the calculated estimated DC value 27a to the comparison unit 28a. [Expression 2] Idce2=(Du1−Du2)×Ius+(Dv1−Dv2)×Ivs+(Dw1−Dw2)×Iws
[0059] Here, expression (2) is maintained by the following idea. Fig. 5 is a direct current Idc flowing into a direct current sensor 12, an average value of the currents Idca between a smoothing capacitor 92 and the power semiconductors. This current Idca is a sum of the currents lu1, lu2, Iv1, Iv2, Iw1 and Iw2 flowing in the power semiconductors 90a, 90g, 90c, 90i, 90e and 90k according to Kirchhoff's law. When the power semiconductors 90a, 90c and 90e are ON, the currents lu1, Iv1 and Iw1 become equal to the alternating currents lu, Iv and Iw respectively. When power semiconductors 90a, 90c, and 90e are OFF, the currents Iu1, Iv1, and Iw1 are 0. Furthermore, when power semiconductors 90g, 90i, and 90k are ON, the currents Iu2, Iv2, and Iw2 are equal to -lu, -Iv, and -Iw, respectively. When power semiconductors 90g, 90i, and 90k are OFF, the currents Iu2, Iv2, and Iw2 are 0.Consequently, the average values of the currents lu1, Iv1, and Iw1 are obtained by multiplying the AC values lu, Iv, and Iw by the ON-time ratios of the power semiconductors 90a, 90c, and 90e, respectively. Furthermore, the average values of the currents Iu2, Iv2, and Iw2 are obtained by multiplying -lu, -Iv, and -Iw by the ON-time ratios of the power semiconductors 90g, 90i, and 90k, respectively. The ON-time ratios of the power semiconductors 90a, 90c, and 90e are the duty cycle values Du1, Dv1, and Dw1, respectively, while the ON-time ratios of the power semiconductors 90g, 90i, and 90k are Du2, Dv2, and Dw2, respectively. Because the AC current values lu, Iv and Iw can be replaced by the AC current sensor values lus, Ivs and Iws, the DC current Idc in the case of the three-phase six-wire system can be estimated by calculating expression (2).
[0060] The comparator unit 28a compares the estimated DC current value Idce2, which is output by the computation unit 26a for the estimated DC current, and a DC current sensor value Idcs and determines, based on the comparison result, whether the DC current sensor 12 is anomalous. In the case where the comparator unit 28a determines that the DC current sensor 12 is anomalous, the comparator unit 28a outputs an anomaly message signal 6 to an anomaly message device 4.
[0061] It is stated that in the present embodiment, the DC sensor diagnostic unit 25a is illustrated within the inverter control device 16a. However, the DC sensor diagnostic unit 25a can be provided outside the inverter control device 16a. Furthermore, the DC sensor diagnostic unit 25a can be mounted in an electronic control device separate from the power conversion device 16a.
[0062] Fig. Figure 6 is a flowchart illustrating the diagnostic processing of the DC sensor 12 in the second embodiment. This diagnostic processing is performed by the inverter control device 16a with an arbitrary time control. It is stated that the parts performing the same processing as the diagnostic processing in the first embodiment are given the same reference numerals, with their descriptions omitted.
[0063] During diagnostic processing in Fig. In step 6, processing is performed in step S110 instead of step S101 of the first embodiment. Furthermore, processing is performed in step S111 instead of step S102 of the first embodiment. Additionally, processing is performed in step S112 instead of step S104.
[0064] In step S110, the calculation unit 26a calculates the estimated DC current value Idce2 based on expression (2) using the duty cycle values Du1, Du2, Dv1, Dv2, Dw1 and Dw2 and the AC sensor values lus, Ivs and Iws.
[0065] Next, in step S111, the comparator unit 28a compares the DC current sensor value Idcs and the estimated DC current value Idce2, determining whether the difference between the DC current sensor value Idcs and the estimated DC current value Idce2 is a threshold of 2 or greater. If the difference is a threshold of 2 or greater, the comparator unit 28a determines that the DC current sensor 12 is anomalous and proceeds to processing in step S103. If the difference is less than a threshold of 2, the inverter control device 16a terminates the diagnostic processing.
[0066] In step S112, the PWM signal generation unit 21a controls the PWM signals 22a to 22l in such a way that the motor 2a is not driven.
[0067] It is stated that the output of the anomaly message signal 6 to the PWM signal generation unit 21a in step S103 and the control of the PWM signals in step S112 are not essential. After the determination that the DC sensor 12 is anomalous has been carried out, only the output of the anomaly message signal 6 to the anomaly message device 4 can be performed.
[0068] Furthermore, during diagnostic processing in Fig. 6. The determination that the DC current sensor 12 is anomalous is made if the difference between the DC current sensor value Idcs and the estimated DC current value Idce2 is the threshold value 2 or greater. However, whether the DC current sensor 12 is anomalous can be determined using a further criterion. For example, if the difference between the DC current sensor value Idcs and the estimated DC current value Idce2 is greater than a difference from the previous diagnosis by a fixed value or more, the determination that the DC current sensor 12 is anomalous can be made.
[0069] As described above, according to the present embodiment, even in the case where the motor 2a and the inverter circuit 9a are of the three-phase six-wire system, the estimated DC value Idce2 is calculated on the basis of the AC sensor values lus, Ivs and Iws and the duty cycle values Du1, Du2, Dv1, Dv2, Dw1 and Dw2, whereby the diagnosis of the DC sensor 12 can be carried out continuously on the basis of the estimated DC value Idce2 and the DC sensor value Idcs.
[0070] It is stated that in the present embodiment, the estimated DC current value Idce2 was calculated using the duty cycle values Du1, Du2, Dv1, Dv2, Dw1, and Dw2. However, the estimated DC current value Idce2 can be calculated using the timing information Cu, Cv, and Cw from the switching of the PWM signals instead of the duty cycle values. The third embodiment
[0071] In the present embodiment, an example of a power conversion device with improved diagnostic accuracy is described in addition to the ability to continuously diagnose an anomaly of a DC sensor.
[0072] Fig. Figure 7 is a graphical representation illustrating a configuration example of a power conversion device and a peripheral circuit in the third embodiment. It is stated that the same elements as in the configuration example in the first embodiment are given the same reference numerals and that the description of these elements is omitted.
[0073] A power conversion device 1b in the third embodiment includes an inverter control device 16b, which differs from that of the first embodiment. The inverter control device 16b includes a DC sensor diagnostic unit 25b, which differs from the DC sensor diagnostic unit 25 in the first embodiment.
[0074] The DC sensor diagnostic unit 25b includes, in addition to the configuration of the DC sensor diagnostic unit 25 in the first embodiment, a correction value calculation unit 29. Furthermore, the DC sensor diagnostic unit 25b includes a comparison unit 28b, which differs from the comparison unit 28 in the first embodiment. Additionally, a function for adding a correction value Ic output by the correction value calculation unit 29 to an estimated DC value Idce1, in order to calculate an estimated DC value after correction Idcc, is added to the DC sensor diagnostic unit 25b.
[0075] The comparison unit 28b compares the DC current sensor value Idcs and, instead of the estimated DC current value Idce1, the estimated DC current value after correction Idcc, and determines from a comparison result whether the DC current sensor 12 is anomalous.
[0076] The correction value calculation unit 29 calculates a correction value 30 according to expression (3) using the AC sensor values lus, Ivs, and Iws. In expression (3), K represents a correction coefficient, where a value greater than 0 and less than 1 is used as the correction coefficient. It is stated that when K is set to 0.5, the error exhibited by the estimated DC value after correction, Idcc, can be reduced the most. The reason is described in the section of the action example that follows. [Expression 3] Ic=−K×(Ius+Ivs+Iws)
[0077] It is stated that in the present embodiment, the DC sensor diagnostic unit 25b is illustrated within the inverter control device 16b. However, the DC sensor diagnostic unit 25b can be provided outside the inverter control device 16b. Additionally, the DC sensor diagnostic unit 25b can be mounted in an electronic control device separate from the power conversion device 16b. Fig. Figure 8 is a flowchart illustrating the diagnostic processing of the DC sensor 12 in the third embodiment. This diagnostic processing is performed by the inverter control device 16b with an arbitrary timer. It is stated that the parts performing the same processing as the diagnostic processing in the first embodiment are given the same reference numerals and that their descriptions are omitted.
[0078] In the diagnostic processing after Fig. In section 8, steps S120 and S121 are added to the diagnostic processing of the first embodiment. Additionally, the processing in step S122 is performed instead of step S102 of the first embodiment.
[0079] In step S120, the correction value calculation unit 29 calculates a correction value Ic based on expression (3) using the AC sensor values lus, Ivs, and Iws. Then, in step S121, the DC sensor diagnostic unit 25b adds the estimated DC value Idce1 and the correction value Ic to calculate the estimated DC value after correction Idcc.
[0080] Next, in step S122, the comparator unit 28b compares the DC current sensor value Idcs and the estimated DC current value after correction Idcc, determining whether the difference between the DC current sensor value Idcs and the estimated DC current value after correction Idcc is a threshold of 3 or greater. If the difference is a threshold of 3 or greater, the comparator unit 28b determines that the DC current sensor 12 is anomalous and proceeds to processing in step S103. If the difference is less than a threshold of 3, the inverter control device 16b terminates the diagnostic processing.
[0081] Furthermore, the diagnostic processing is carried out according to Fig. 8. The determination that the DC current sensor 12 is anomalous is made if the difference between the DC current sensor value Idcs and the estimated DC current value after correction Idcc is a threshold of 3 or greater. However, whether the DC current sensor 12 is anomalous can also be determined using a further criterion. If the difference between the DC current sensor value Idcs and the estimated DC current value after correction Idcc is greater than a fixed value or more than a difference in the previous diagnosis, the determination that the DC current sensor 12 is anomalous can be made.
[0082] Fig. Figure 9 is an example illustrating the effect of the correction in the case where the AC sensor value lus of the U-phase contains an offset error loff. The upper curve in Fig. Figure 9 illustrates the duty cycle values Du, Dv, and Dw of the U-phase, V-phase, and W-phase, respectively, while the middle curve illustrates the AC sensor values lus, Ivs, and Iws of the U-phase, V-phase, and W-phase, and a three-phase sum of the AC sensor values. The bottom curve illustrates an actual DC value, an estimated DC value Idce1 before correction, and an estimated DC value after correction Idcc. The correction value Ic is calculated based on expression (3), where in Fig. 9 the correction coefficient K 0.5. Furthermore, the horizontal axes represent an electrical angle of motor 2.
[0083] In Fig. 9. The actual DC value has a fixed value. Because the AC sensor value lus of the U-phase has the offset error loff, an error of up to loff occurs in the estimated DC value Idce1 compared to the actual DC value. Furthermore, in the case of determining an average value of the estimated DC values Idce1 in one cycle of the electric angle, the average of the estimated DC values Idce1 has an error of loff × 0.5 compared to the actual DC value. In contrast, the error between the estimated DC value after correction Idcc and the actual DC value is reduced to loff × 0.5. Furthermore, an average of the estimated DC values after correction Idcc is equal to the actual DC value.
[0084] Fig. Figure 10 illustrates the amount of error that the estimated DC value has after correction Idcc, in the case where the value of the correction coefficient K in the example after Fig. 9 is changed. In the example after Fig. 9. The three-phase sum of the AC sensor values lus, Ivs, and Iws is calculated as loff. Therefore, the correction value Ic is increased by -0.1 × loff when the correction coefficient K is increased by 0.1. As a result, the estimated DC value after correction, Idcc, is reduced by 0.1 × loff.
[0085] Out of Fig. 10. If the correction coefficient K is 1, the magnitude of the error in the estimated DC value after correction Idcc will be the same as in a case where the correction coefficient K is 0 (= no correction). Furthermore, if the correction coefficient K is 0.5, the magnitude of the error in the estimated DC value after correction Idcc will be smallest.
[0086] Fig. Figure 11 is an example illustrating the effect of the correction in the case where the AC sensor value lus of the U-phase and the AC sensor value Ivs of the V-phase exhibit the offset error loff. The arrangement of the curves in Fig. 11 is to the in Fig. 9 similar, whereby in Fig. 11 the value of the correction coefficient K is 0.5.
[0087] In Fig. 11. Compared to the actual DC value, an error of up to loff × 1.5 occurs in the estimated DC value Idce1. Furthermore, when determining the average of the estimated DC values Idc1 in one cycle of the electric angle, the average of the estimated DC values Idce1 exhibits an error of loff compared to the actual DC value. In contrast, the error between the estimated DC value after correction Idcc and the actual DC value is reduced to loff × 0.5. Furthermore, the average of the estimated DC values after correction Idcc is equal to the actual DC value.
[0088] Fig. Figure 12 illustrates the amount of error that the estimated DC value has after correction Idcc, in the case where the value of the correction coefficient K in the example after Fig. 11 is changed. In the example after Fig. 11. The three-phase sum of the AC sensor values lus, Ivs, and Iws is loff × 2. Therefore, the correction value Ic is increased by -0.2 × loff when the correction coefficient K is increased by 0.1. As a result, the estimated DC value after correction, Idcc, is decreased by 0.2 × loff.
[0089] In the example according to Fig. 11. If the correction coefficient K is 1, the magnitude of the error will be the same as in the case where the correction coefficient K is 0. Furthermore, if the correction coefficient K is 0.5, the magnitude of the error of the estimated DC value after correction Idcc will be smallest.
[0090] Fig. Figure 13 is an example illustrating the effect of the correction in the case where the AC sensor value lus of the U-phase, the AC sensor value Ivs of the V-phase, and the AC sensor value Iws of the W-phase have the offset error loff. The arrangement of the curves in Fig. 13 is to the in Fig. 9 similar, whereby in Fig. 13 the value of the correction coefficient K is 0.5.
[0091] In the example according to Fig. 13. The estimated DC value Idce1, regardless of the electrical angle, has an error of loff × 1.5 compared to the actual DC value. In contrast, after correction, the DC value Idcc equals the actual DC value.
[0092] Fig. Figure 14 illustrates the amount of error that the estimated DC value has after correction Idcc in the case where the value of the correction coefficient K in the example after Fig. 13 is changed. In the example after Fig. 13. The three-phase sum of the AC sensor values lus, Ivs, and Iws is loff × 3. Therefore, the correction value Ic is increased by -0.3 × loff when the correction coefficient K is increased by 0.1. As a result, the estimated DC value after correction, Idcc, is reduced by 0.3 × loff.
[0093] If in the example after Fig. 13. If the correction coefficient K is 1, the magnitude of the error will be the same as in the case where the correction coefficient K is 0. Furthermore, if the correction coefficient K is 0.5, the magnitude of the error of the estimated DC value after correction Idcc will be smallest.
[0094] Fig. Figure 15 is an example illustrating the effect of correction in the case where the AC sensor reading of the U-phase has a gain error Ig. The arrangement of the curves in Fig. 15 is to the in Fig. 9 similar, whereby in Fig. 15 the value of the correction coefficient K is 0.5. Because the AC sensor of the U-phase has the gain error Ig, the value of lus in the middle curve is increased by lu × Ig when a primary AC value of the U-phase is lu. Because in this example a maximum value of the primary AC value of the U-phase is Imax, a maximum error value of lus is Imax × Ig.
[0095] In the example according to Fig. 15. Compared to the actual DC value, an error of up to Imax × Ig occurs in the estimated DC value Idce1. Furthermore, when determining an average value of the estimated DC values Idce1 in one cycle of the electric angle, the average of the estimated DC values Idce1 has an error of Imax × Ig × 0.25 compared to the actual DC value. In contrast, the error between the estimated DC value after correction Idcc and the actual DC value is reduced to Imax × Ig × 0.5. However, an average value of the estimated DC values after correction Idcc will be the same as an average value of the estimated DC values Idce1 before correction, with the error of the average value not being reduced.
[0096] Fig. Figure 16 illustrates the amount of error that the estimated DC value has after correction Idcc in the case where the value of the correction coefficient K in the example after Fig. 15 is changed. In the example after Fig. 15. The three-phase sum of the AC sensor values lus, Ivs and Iws is lu × Ig. Therefore, the correction value Ic is increased by -0.1 × lu × Ig when the correction coefficient K is increased by 0.1.
[0097] If in the example after Fig. 15. If the correction coefficient K is 1, the magnitude of the error will be the same as in the case where the correction coefficient K is 0. Furthermore, if the correction coefficient K is 0.5, the magnitude of the error of the estimated DC value after correction Idcc will be smallest. It is stated that the error exhibited by the average value of the estimated DC values after correction Idcc is constant regardless of the value of the correction coefficient K.
[0098] As described above, according to the present embodiment, the correction value calculation unit 29 calculates the correction value Ic using the AC sensor values lus, Ivs, and Iws, wherein the comparison unit compares the estimated DC value after correction Idcc and the DC sensor value Idcs to perform the diagnosis. Because the deviation of the estimated DC value Idce1, which is caused by the error exhibited by the AC sensor values 14a to 14c, can be reduced by the correction value Ic, the anomaly detection threshold 3 in the present embodiment can be set to be lower than the threshold 1 of the first embodiment.As a result, in the case where the DC sensor value Idcs deviates from the primary value, the anomaly can be detected even in the case of a small deviation, whereby the diagnosis of the DC sensor 12 can be carried out more accurately.
[0099] Furthermore, from the into the Fig. The examples illustrated in Figures 9 to 16 show that the error from the actual DC value can be further reduced by determining the average of the estimated DC values after correction Idcc. Therefore, the diagnosis of the DC sensor 12 can be performed by averaging the estimated DC values after correction Idcc using a low-pass filter or the like, and by using an averaged estimated DC value after correction and the DC sensor value Idcs.
[0100] It is stated that the need for the DC current sensor 12 is high when diagnosing the power conversion device. However, from a cost standpoint, there is a challenge to eliminating the DC current sensor 12. The calculation of the estimated DC current and the calculation of the correction value, as described in the present embodiment, can also be used to address the challenge of eliminating the DC current sensor 12. By accurately estimating the DC current value using the methods described in the present embodiment, the cost can be reduced by eliminating the DC current sensor 12, while maintaining an accuracy equivalent to that achieved when the DC current sensor 12 is included.
[0101] The present invention is not limited to the embodiments described above and includes various modifications. The embodiments described above have been explained in detail, for example, to facilitate understanding of the present invention, which is not necessarily limited to an embodiment containing all the described configurations. Furthermore, part of the configuration of a particular embodiment may be replaced by the configuration of another embodiment. Furthermore, the configuration of another embodiment may be added to the configuration of a particular embodiment. Furthermore, for some of the configurations of the embodiments, a further configuration may be added, deleted, or replaced.Furthermore, some or all of the configurations, functions, processing units, and processing means described above may be implemented by hardware designed with an integrated circuit or the like. Alternatively, the configurations, functions, and the like described above may be implemented by software in such a way that the programs implementing the respective functions are interpreted and executed by the processor. The information, such as the programs, tables, and files implementing the functions, may be stored in a recording device, such as a memory, hard disk drive, or solid-state drive (SSD), or in a recording medium, such as an IC card, SD card, or DVD. List of reference symbols 1 Power conversion device 1a Power conversion device 1b Power conversion device 5 Target torque 6 Anomaly alert signal 9 Inverter circuit 9a Inverter circuit 12 DC sensor 14a AC sensor 14b AC sensor 14c AC sensor 16 Inverter control device 16a Inverter control device 16b Inverter control device 22a PWM signal 22b PWM signal 22c PWM signal 22d PWM signal 22e PWM signal 22f PWM signal 22g PWM signal 22h PWM signal 22i PWM signal 22j PWM signal 22k PWM signal 22l PWM signal 25 DC sensor diagnostic unit 25a DC sensor diagnostic unit 25b DC sensor diagnostic unit 26 units of calculation for the estimated direct current 26a Unit of calculation for the estimated direct current 28 comparison units 28a Comparison unit 28b Comparison unit 29 Correction value calculation unit Idcc estimated DC value after correction Your duty cycle value Dv duty cycle value Dw duty cycle value Du1 Duty cycle value Du2 duty cycle value Dv1 Duty cycle value Dv2 Duty cycle value Dw1 Duty cycle value Dw2 Duty cycle value Ic correction value Idce1 estimated DC value Idce2 estimated DC value IDC DC sensor value plus AC sensor value IVs AC sensor value IWS AC sensor value
Claims
[1] Inverter control device (16) controlling an inverter circuit (9), wherein the inverter control device (16) is configured as follows: to calculate an estimated DC current value based on a duty cycle value and an AC current sensor value output by an AC current sensor, to perform a diagnosis of a DC current sensor (12) based on the estimated DC current value and a DC current sensor value output by the DC current sensor (12), and to correct the estimated DC value based on a sum of the AC values of the AC sensor values and to perform the diagnosis of the DC sensor (12) based on an estimated DC value after correction and the DC sensor value. [2] Inverter control device (16) according to claim 1, which is further configured to perform filtering processing for the estimated DC value after correction and to perform the diagnosis of the DC sensor (12) based on an estimated DC value after filtering and the DC sensor value. [3] Inverter control device (16) according to claim 1 or 2, which is further configured to output an anomaly message signal when it is determined that the DC sensor (12) is anomalous. [4] Inverter control device (16) according to claim 3, which is further configured, to control the inverter circuit (9) in such a way that a motor (2) is driven according to a target torque, and to control the inverter circuit (9) in such a way that the motor (2) is not driven when it is determined that the DC sensor (12) is anomalous. [5] Power conversion device (1) comprising: an inverter circuit (9); an AC current sensor; a DC current sensor (12); and an inverter control unit, wherein the inverter control unit controls the inverter circuit (9) in such a way that a motor (2) is driven according to a target torque, and calculates an estimated DC value based on a duty cycle value and the AC current sensor value and performs a diagnosis of the DC current sensor (12) based on the estimated DC current value and a DC current sensor value output by the DC current sensor (12), wherein the inverter control unit corrects the estimated DC value based on a sum of the AC values of the AC sensor values and performs the diagnosis of the DC sensor (12) based on an estimated DC value after correction and the DC sensor value. [6] Power conversion device (1) according to claim 5, wherein the inverter control unit performs filtering processing for the estimated DC value after correction and performs the diagnosis of the DC sensor (12) based on an estimated DC value after filtering and the DC sensor value. [7] Power conversion device (1) according to claim 5 or 6, wherein the inverter control unit outputs an anomaly message signal when it is determined that the DC sensor (12) is anomalous. [8] Power conversion device (1) according to claim 7, wherein the inverter control unit controls the inverter circuit (9) in such a way that the motor (2) is not driven when it is determined that the DC sensor (12) is anomalous.
Citation Information
Patent Citations
Electric power steering
DE102013217970A1
Motor drive control system
JP5760778B2
JP000005760778B2