Anomaly detection device

Abstract

To provide an abnormality detection device capable of precisely detecting overcurrent abnormality due to ON fixation of an inverter switching element.SOLUTION: Amplifiers (monitor voltage output units) 501, 502 and 503 output monitor voltages Vau, Vav and Vaw having positive correlation with currents flowing to shunt resistances 67, 68 and 69 connected in series to upper arm elements 61, 62 and 63 and lower arm elements 64, 65 and 66 of respective phases. A monitor circuit 40 monitors inverter overcurrent abnormality, i.e., flowing of a through current due to ON fixation of the upper arm elements or the lower arm elements of one phase or more. A comparator of the monitor circuit 40 outputs an overcurrent flag of a pulse signal when a monitor voltage is larger than a reference value. A determination part determines the inverter overcurrent abnormality and stops feeding a current to an inverter 60 when a cumulative output time of the overcurrent flag of at least one phase reaches a time threshold or when the frequency of output of the overcurrent flag reaches a frequency threshold.SELECTED DRAWING: Figure 1

Description

[Technical field]

[0001] The present invention relates to an abnormality detection device. [Background technology]

[0002] 2. Description of the Related Art Abnormality detection devices that detect ON-fixed abnormality (short circuit failure) of inverter switching elements are known.

[0003] For example, the device disclosed in Patent Document 1 detects a short circuit fault in an upper arm element or a lower arm element based on a divided voltage at a connection point between an upper arm element and a lower arm element of each phase. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] JP 2018-113851 A Summary of the Invention [Problem to be solved by the invention]

[0005] In the device of Patent Document 1, the divided voltage at the connection point varies due to the ON resistance of the switching element (FET) or the like, which may lead to erroneous detection.

[0006] The present invention has been made in view of the above-mentioned points, and has an object to provide an abnormality detection device capable of accurately detecting an overcurrent abnormality caused by an inverter switching element being stuck in the ON state. [Means for solving the problem]

[0007] The abnormality detection device of the present invention includes a control unit (30), a plurality of monitored voltage output devices (501, 502, 503), and a monitoring circuit (40). The control unit causes upper arm elements (61, 62, 63) and lower arm elements (64, 65, 66) of multiple phases in an inverter (60) in which the upper arm elements and the lower arm elements are bridge-connected to perform complementary switching operations.

[0008] The monitor voltage output device outputs a monitor voltage (Vau, Vav, Vaw) having a positive correlation with the current flowing through a shunt resistor (67, 68, 69) connected in series with the upper arm element and the lower arm element of each phase. The monitor circuit monitors the inverter overcurrent abnormality caused by the upper arm element or the lower arm element of one or more phases being stuck ON based on the monitor voltage during inverter operation. The monitor voltage generator holds the output state of the monitor voltage for a predetermined hold time from the input of the trigger signal.

[0009] The monitoring circuit includes comparators (411, 412, 413) and a determination unit (45). When the monitored voltage is greater than a reference value (Vref), the comparator outputs an overcurrent flag as a pulse signal.

[0010] In a first aspect of the present invention, the determination unit includes: When the cumulative output time (ΣToc) of at least one phase of the overcurrent flag reaches the time threshold (Tth). 、 It is determined that an inverter overcurrent abnormality has occurred, and power to the inverter is stopped. In a second aspect, the judgment unit judges that an inverter overcurrent abnormality has occurred when the cumulative output time (ΣToc) of the overcurrent flag of at least one phase reaches a time threshold (Tth) or when the number of times the overcurrent flag is output (Noc) reaches a number threshold (Nth), whichever occurs earlier, and stops the flow of current to the inverter.

[0011] In the present invention, by using an overcurrent flag generated based on a monitoring voltage that has a positive correlation with the current flowing through the shunt resistor, it is possible to accurately detect inverter overcurrent abnormalities without being affected by variations in ON resistance as in the conventional technology of Patent Document 1. [Brief description of the drawings]

[0012] [Figure 1] FIG. 1 is a configuration diagram of a three-phase motor drive device to which an abnormality detection device according to each embodiment is applied. [Diagram 2] FIG. 2 is a block diagram of a monitoring circuit according to an example configuration. [Diagram 3] FIG. 13 is a block diagram of a monitoring circuit according to another example configuration. [Figure 4] A characteristic diagram of the monitored voltage (amplifier output) versus the current flowing through the shunt resistor. [Diagram 5] 5 is a time chart illustrating a monitored voltage and an overcurrent flag when an upper arm element or a lower arm element is abnormally stuck ON. [Figure 6] 4 is a time chart showing inverter overcurrent anomaly detection according to the first embodiment. [Figure 7] 6 is a time chart showing a case where the output of a monitored voltage is maintained. [Figure 8] 4 is a flowchart of an initial check of a monitoring circuit. [Figure 9] 4 is a flowchart of inverter overcurrent anomaly detection according to the first embodiment. [Figure 10] 10 is a time chart showing inverter overcurrent anomaly detection according to the second embodiment. [Figure 11] 10 is a flowchart of inverter overcurrent anomaly detection according to a second embodiment. [Figure 12] 13 is a flowchart of inverter overcurrent anomaly detection according to the third embodiment. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0013] Abnormality detection devices according to multiple embodiments of the present invention will be described with reference to the drawings. The abnormality detection device of the present embodiment detects an inverter overcurrent abnormality caused by a switching element stuck on in an inverter that converts DC power from a battery and supplies it to a load such as a motor. The abnormality detection device of the present embodiment is applied to, for example, a motor drive device that drives a steering assist motor of an electric power steering device.

[0014] 1 shows a schematic configuration of a motor drive device 10 to which an abnormality detection device 20 is applied. DC power from a battery 15 is converted into three-phase AC power by the switching operation of an inverter 60 and supplied to a motor 80. The motor 80 is a three-phase brushless motor.

[0015] In the inverter 60, three-phase upper and lower arm switching elements 61-66 are bridge-connected between a power supply line Lp and a ground line Lg. In this embodiment, MOSFETs are used as the switching elements 61-66. Hereinafter, the upper arm switching elements are referred to as "upper arm elements" and the lower arm switching elements are referred to as "lower arm elements". The upper arm element 61 and the lower arm element 64 form a U-phase leg, the upper arm element 62 and the lower arm element 65 form a V-phase leg, and the upper arm element 63 and the lower arm element 66 form a W-phase leg. The arm connection points Nu, Nv, and Nw of each phase are connected to the motor winding. A motor relay may be provided between the arm connection points Nu, Nv, and Nw and the motor winding.

[0016] Shunt resistors 67, 68, 69 are connected in series to the upper arm elements 61, 62, 63 and the lower arm elements 64, 65, 66 of each phase. In the configuration example of Fig. 1, the shunt resistors 67, 68, 69 are connected to the low potential sides of the lower arm elements 64, 65, 66. Furthermore, a smoothing capacitor 55 formed of an aluminum electrolytic capacitor or the like is connected in parallel with the upper and lower arm elements of each phase between the power supply line Lp and the ground line Lg.

[0017] A reverse connection protection relay 17 is provided in the power supply line Lp from the battery 15 to the inverter 60. When the reverse connection protection relay 17 is OFF, it cuts off the current from the inverter 60 side to the battery 15 side. Note that a power supply relay may be further provided between the battery 15 and the reverse connection protection relay 17, which cuts off the current from the battery 15 side to the inverter 60 side when it is OFF.

[0018] Incidentally, when any of the switching elements 61-66 of the inverter 60 is stuck ON, a through current flows from the power supply line Lp to the ground line Lg during the ON period of the normal element that forms an upper or lower pair with that element. For example, an excessive through current may flow due to the charge stored in the smoothing capacitor 55 being instantly released. In this specification, an abnormality in which a through current flows due to an upper arm element or lower arm element of one or more phases being stuck ON is defined as an "inverter overcurrent abnormality." 。

[0019] For example, in the conventional technology of Patent Document 1 (JP Patent Publication No. 2018-113851), the divided voltages at the inter-arm connection points Nu, Nv, and Nw may vary due to the ON resistance of the switching elements, which may lead to erroneous detection. Therefore, the abnormality detection device 20 of the present embodiment aims to prevent the occurrence of malfunctions by accurately detecting inverter overcurrent abnormalities and stopping the current supply to the inverter 60 early.

[0020] The abnormality detection device 20 includes a control unit 30, a monitoring circuit 40, and three amplifiers 501, 502, and 503 serving as "monitoring voltage output devices."

[0021] The control unit 30 is composed of a microcomputer, a pre-driver, etc., and includes a CPU, ROM, RAM, I / O, and bus lines connecting these components (not shown). The control unit 30 executes software processing by having the CPU execute a program stored in advance in a substantial memory device (i.e., a readable non-transitory tangible recording medium) such as a ROM, or performs control through hardware processing using a dedicated electronic circuit. The monitoring circuit 40 is composed of, for example, a customized IC.

[0022] The control unit 30 causes the upper arm elements 61, 62, 63 and the lower arm elements 64, 65, 66 of the inverter 60 to perform complementary switching operations. During normal operation when the inverter 60 is normal, the control unit 30 uses a general motor control technique such as vector control or current feedback control to issue a drive signal to the inverter 60 so as to cause the motor 80 to output the required torque. In addition, the control unit 30 operates the reverse connection protection relay 17, etc.

[0023] The amplifiers 501, 502, and 503 output monitoring voltages Vau, Vav, and Vaw having a positive correlation with the current flowing through the shunt resistors 67, 68, and 69 of each phase to the monitoring circuit 40. Specific characteristics of the monitoring voltages Vau, Vav, and Vaw with respect to the current will be described later with reference to Fig. 4. Also, the trigger signal input from the control unit 30 to the amplifiers 501, 502, and 503 of each phase will be described later with reference to Fig. 7.

[0024] The monitoring circuit 40 monitors an inverter overcurrent anomaly based on the monitoring voltages Vau, Vav, and Vaw input from the amplifiers 501, 502, and 503. If it is determined that there is an inverter overcurrent anomaly, the monitoring circuit 40 turns off all of the inverter switching elements 61-66 to stop the passage of current through the inverter 60. Note that the possibility of a double fault in which the upper arm element and the lower arm element of the same phase are stuck ON at the same time is not considered. Therefore, when either the upper arm element or the lower arm element is stuck ON, a shoot-through current can be prevented by keeping the other element constantly OFF.

[0025] Furthermore, when the monitoring circuit 40 determines that an inverter overcurrent abnormality has occurred, it notifies the control unit 30 of the abnormality. Furthermore, in an initial check of the monitoring circuit 40, which will be described later, the control unit 30 diagnoses whether the monitoring circuit 40 functions normally when a monitoring voltage for diagnosis is input.

[0026] 2 shows a block diagram of a configuration example of a monitoring circuit 40A. The monitoring circuit 40 has comparators 411, 412, 413, filters 421, 422, 423, accumulating counters 431, 432, 433, and latch circuits 441, 442, 443 for each phase. The numbers "1", "2", and "3" at the end of the reference numbers correspond to the U phase, V phase, and W phase, respectively. The monitoring circuit 40 also has a determination unit 45 that is common to the three phases. Until the input to the determination unit 45, the same processing is performed in parallel for each phase.

[0027] The comparators 411, 412, 413 output an overcurrent flag as a pulse signal when the monitoring voltages Vau, Vav, Vaw input from the amplifiers 501, 502, 503 are greater than a reference value Vref. The filters 421, 422, 423 remove noise when the overcurrent flag is ON or OFF as a high-frequency component superimposed on the overcurrent flag. The accumulating counters 431, 432, 433 count the accumulated output time or the number of outputs of the overcurrent flag, as will be described in detail later. The latch circuits 441, 442, 443 hold the count values ​​of the accumulating counters 431, 432, 433.

[0028] The count values ​​of the integrating counters 431, 432, 433 and the latch circuits 441, 442, 443 are periodically cleared at the end of each monitoring period, as described later with reference to Fig. 6 and Fig. 10. This prevents momentary erroneous detection results from accumulating and causing abnormality determination. The count values ​​are also cleared after being intentionally counted up during an initial check of the monitoring circuit 40, which will be described later.

[0029] The determination unit 45 compares the count values ​​held in the latch circuits 441, 442, 443 of each phase with a threshold value, and determines whether the count value is equal to or greater than the threshold value by OR operation. In other words, when the count value is equal to or greater than the threshold value in at least one phase, the determination unit 45 determines that an inverter overcurrent abnormality has occurred, and stops the supply of current to the inverter 60.

[0030] Fig. 3 shows a block diagram of a monitoring circuit 40B of another configuration example. Components substantially the same as those in Fig. 2 are given the same reference numerals and will not be described. In contrast to the monitoring circuit 40A in Fig. 2, the monitoring circuit 40B in Fig. 3 has one set of integrating counter 43 and latch circuit 44 common to the three phases. The integrating counter 43 counts the cumulative output time or number of outputs for the three phases in total for the overcurrent flags input from the filters 421, 422, and 423 of each phase. The latch circuit 44 holds the count value for the three phases in total by the integrating counter 43. The other configurations of the monitoring circuit 40B are the same as those of the monitoring circuit 40A.

[0031] When the total count value of the three phases is equal to or greater than the threshold value, the determination unit 45 determines that the count value in at least one phase is equal to or greater than the threshold value, determines that an inverter overcurrent abnormality has occurred, and stops the supply of current to the inverter 60. The monitoring circuit 40B can reduce the mounting area by reducing the number of integrating counters and latch circuits compared to the monitoring circuit 40A.

[0032] Hereinafter, for matters common to the configuration examples of FIG. 2 and FIG. 3, the reference symbols of the monitoring circuits 40A and 40B will not be distinguished and will be referred to as "monitoring circuit 40". In addition, the three-phase accumulating counters 431, 432, and 433 and the latch circuits 441, 442, and 443 in the configuration example of FIG. 2 will be collectively referred to as "accumulating counter 43" and "latch circuit 44". The accumulating counter 43 in the configuration example of FIG. 2 counts the accumulated output time or the number of outputs of the overcurrent flag of each phase, and the latch circuit 44 holds the counter value. The accumulating counter 43 in the configuration example of FIG. 3 counts the accumulated output time or the number of outputs of the overcurrent flag of the total of the three phases, and the latch circuit 44 holds the counter value.

[0033] As described above, the monitoring circuit 40 can prevent malfunctions caused by continuing to energize the inverter in an overcurrent abnormality state, and can improve the reliability of the system. However, if the monitoring circuit 40 is stuck on the normal judgment side, there is a problem that the monitoring function is lost. Therefore, in this embodiment, the control unit 30 performs an initial check of the monitoring circuit 40.

[0034] In an initial check of the monitoring circuit 40, the control unit 30 causes the amplifiers 501, 502, 503 of each phase to output a voltage larger than a reference value as a monitoring voltage for diagnosis to the comparators 411, 412, 413. For example, a monitoring voltage for diagnosis of 5V is output for a comparator reference value Vref of 4V. The control unit 30 obtains the output signals from the filters 421, 422, 423 to the integrating counter 43, and diagnoses whether the overcurrent flag is being output correctly. This is called a "comparator check."

[0035] As described later, when the cumulative output time or the number of outputs of the diagnostic monitoring voltage meets the abnormality determination condition, the control unit 30 diagnoses that the determination unit 45 determines that an inverter overcurrent abnormality has occurred. Specifically, it is determined that an overcurrent abnormality notification flag is sent from the monitoring circuit 40 to the control unit 30. This is called a "logic check."

[0036] 4 shows the characteristics of the monitoring voltage Va (amplifier output) versus the current I flowing through the shunt resistors 67, 68, 69. The monitoring voltage Va is a term that collectively includes the monitoring voltages Vau, Vav, and Vaw of each phase. The resistance value of the shunt resistors 67, 68, 69 is Rs, the gain of the amplifiers 501, 502, 503 is G (≧1), and the amplifier output when the current I is 0 is an offset voltage Vofs. The monitoring voltage Va has a linear characteristic expressed by the formula "Va=Rs×I×G+Vofs" in the range from the lower limit value Va_min to the upper limit value Va_max of the amplifier output.

[0037] The offset voltage Vofs is set so that the monitoring voltage Va becomes positive even when a negative current I (<0) flows from the low potential side to the high potential side of the shunt resistors 67, 68, 69. Note that in this embodiment, a situation is basically assumed in which the current I is 0 or a positive current I flows from the high potential side to the low potential side of the shunt resistors 67, 68, 69.

[0038] The comparator reference value Vref is set to a value greater than the offset voltage Vofs and less than the monitoring voltage upper limit value Va_max. The current region on the characteristic line that is less than the reference current Iref corresponding to the comparator reference value Vref is considered to be a normal current region that may flow during normal operation. The current region that is greater than or equal to the reference current Iref is determined to be an overcurrent abnormality region.

[0039] With reference to the time chart in Figure 5, the changes in the monitoring voltage Va and the overcurrent flag when the upper arm element or the lower arm element has an ON stuck abnormality will be described. Hereinafter, the duty ratio means the ratio of the ON period of the upper arm element to the switching period. When the duty ratio is larger than the PWM carrier, the gate command for the upper arm element is ON and the gate command for the lower arm element is OFF. When the upper arm element has an ON stuck abnormality, a through current flows during the period when the gate command for the upper arm element is OFF, and the monitoring voltage Va exceeds the comparator reference value Vref, so the overcurrent flag is turned ON. The time during which the overcurrent flag is output as ON is represented as the output time Toc. Similarly, when the lower arm element has an ON stuck abnormality, a through current flows during the period when the gate command for the lower arm element is OFF, and the monitoring voltage Va exceeds the comparator reference value Vref, so the overcurrent flag is turned ON.

[0040] However, since the overcurrent flag may accidentally be turned on due to noise or the like, it is not appropriate to determine an abnormality based on a single overcurrent flag. Therefore, the monitoring circuit 40 of this embodiment determines that an inverter overcurrent abnormality has occurred when the cumulative output time or number of outputs of the overcurrent flag becomes equal to or exceeds a threshold value. Details of this will be described below in the order of the first to third embodiments.

[0041] (First embodiment) FIG. 6 shows a time chart of inverter overcurrent abnormality detection according to the first embodiment. The integrating counter 43 counts the accumulated output time ΣToc obtained by accumulating the output time Toc of the overcurrent flag for each time. When the overcurrent flag is ON, the counter value is incremented, and when the overcurrent flag is OFF, the counter value does not change. For example, when the accumulated output time ΣToc reaches the time threshold value Tth at time tx during the sixth output of the overcurrent flag, the determining unit 45 determines that an inverter overcurrent abnormality has occurred, and stops the supply of electricity to the inverter 60. When a predetermined monitoring period has elapsed since the start of the integration, the integrating counter 43 is periodically cleared.

[0042] However, when the upper arm element is stuck ON when the duty ratio is close to 0%, or when the lower arm element is stuck ON when the duty ratio is close to 100%, the time during which the through current flows becomes shorter, and the output time Toc of each overcurrent flag becomes shorter. This results in a problem that it takes a longer time for the cumulative output time ΣToc to reach the time threshold Tth, slowing down the detection of an abnormality.

[0043] 7, the output states of the monitoring voltages Vau, Vav, Vaw of the amplifiers 501, 502, 503 are held for a predetermined hold time from input of the trigger signal. The trigger signal is output from the control unit 30 to each of the amplifiers 501, 502, 503 when an overcurrent is detected at the timing of the peak or valley of the PWM carrier, for example. During a period in which the duty ratio is constant, the center of the pulses of the gate signals of the upper arm element and the lower arm element becomes the output timing of the trigger signal.

[0044] 7, the dashed double-dashed line indicates the case where the monitoring voltage Va is not held, and the solid line indicates the case where the monitoring voltage Va is held. The hold time Th from the input of the trigger signal is set to, for example, about 10 μs, which corresponds to the sample hold time of the current sensor.

[0045] In the example where the switching period is 50 μs and the duty ratio is 10%, the output time Toc of the overcurrent flag is 5 μs when the monitored voltage Va is not held. When the monitored voltage Va is held, the trigger signal is input (5 / 2) = 2.5 μs after the overcurrent flag turns ON, and if the hold time Th is 10 μs, the output time Toc of the overcurrent flag is 12.5 μs. Therefore, the time until the accumulated output time ΣToc reaches the time threshold Tth can be reduced by 40%.

[0046] 8 shows a flowchart of the initial check of the monitoring circuit. In the following explanation of the flowchart, the symbol "S" indicates a step. In S01, the amplifiers 501, 502, and 503 output a voltage higher than the reference value Vref as a monitoring voltage for diagnosis to the comparators 411, 412, and 413 of each phase. For example, the amplifiers 501, 502, and 503 output a monitoring voltage for diagnosis of 5V for a time period equal to or longer than the time threshold value Tth with respect to the comparator reference value Vref of 4V, thereby intentionally generating an overcurrent flag.

[0047] In S02, a comparator check is performed to determine whether the comparators 411, 412, and 413 of each phase output an overcurrent flag. If the overcurrent flag is correctly output, a determination of YES is made in S02, and the process proceeds to S03. In the first embodiment, the cumulative output time of the diagnostic monitoring voltage reaches the time threshold value Tth, and thus the abnormality determination condition in S03 is established.

[0048] In S04, as a logic check, the determination unit 45 determines that an inverter overcurrent abnormality has occurred and outputs an overcurrent abnormality notification flag. If the control unit 30 correctly receives the overcurrent abnormality notification flag, YES is determined in S04. As a result, the monitoring circuit 40 is determined to be normal in S05. In S06, the count values ​​of the integrating counter 43 and the latch circuit 44 are cleared, the initial check is completed, and normal operation can be started.

[0049] On the other hand, if the comparator check does not output the overcurrent flag, NO is determined in S02, and if the control unit 30 does not receive the overcurrent abnormality notification flag in the logic check, NO is determined in S04. In these cases, the monitoring circuit 40 is determined to be abnormal in S07, and abnormality measures such as prohibiting normal operation or issuing an alarm are executed.

[0050] 9 shows a flowchart of inverter overcurrent anomaly detection according to the first embodiment. In the flowcharts of inverter overcurrent anomaly detection according to the first to third embodiments, substantially the same steps are given the same step numbers and descriptions thereof will be omitted.

[0051] In S11, the amplifiers 501, 502, 503 of each phase hold the output state of the monitoring voltages Vau, Vav, Vaw for a predetermined hold time Th from the input of the trigger signal. In S12, the integrating counter 43 counts the cumulative output time ΣToc of the overcurrent flag for each phase or for the total of the three phases, and the count value is held by the latch circuit 44. In S13, it is determined whether the cumulative output time ΣToc of the overcurrent flag has reached the time threshold value Tth.

[0052] If the result of S13 is YES, the determination unit 45 determines in S14 that an inverter overcurrent abnormality has occurred, and in S15 stops the passage of current to the inverter 60. Here, the control unit 30 or the monitoring circuit 40 may be configured to be able to identify an abnormal phase in which the upper arm element or the lower arm element is stuck ON, based on the outputs of the comparators 411, 412, 413 for each phase.

[0053] If the answer is NO in S13, the process proceeds to S16, where it is determined whether the monitoring period has elapsed. If the monitoring period has not yet elapsed, the result is NO in S16, and the processes of S11 to S13 are repeated. If the monitoring period has elapsed, the result is YES in S16, and the process proceeds to S17. In S17, the determination unit 45 determines that the inverter 60 is provisionally normal. "Provisionally normal" means that no abnormality was detected during that monitoring period. In S18, the count values ​​of the integrating counter 43 and the latch circuit 44 are cleared (i.e., periodically cleared). After that, the entire routine is returned, and overcurrent abnormality detection during inverter operation continues.

[0054] As described above, in this embodiment, by using the overcurrent flag generated based on the monitored voltages Vau, Vav, and Vaw, it is possible to accurately detect an inverter overcurrent abnormality without being affected by variations in ON resistance as in the conventional technology of Patent Document 1.

[0055] Second embodiment Next, inverter overcurrent abnormality detection according to the second embodiment will be described with reference to Fig. 10 and Fig. 11. In the second embodiment, the cumulative counter 43 of the monitoring circuit 40 counts the number of times Noc the overcurrent flag is output. When the number of times Noc the output is equal to or greater than the number threshold Nth, the determination unit 45 determines that an inverter overcurrent abnormality has occurred.

[0056] 10, the integrating counter 43 detects and counts the rising edge of the overcurrent flag. For example, if the number threshold Nth is set to 5, it is determined that an inverter overcurrent abnormality occurs at time tx when the fifth rising edge of the overcurrent flag is detected from the start of counting.

[0057] In the flowchart of the second embodiment shown in Fig. 11, unlike the first embodiment, S11 in Fig. 9 is eliminated, and S22 and S23 are included instead of S12 and S13. In S22, the integrating counter 43 counts the number of times Noc the overcurrent flag is output for each phase or for the total of the three phases, and the latch circuit 44 holds the count value. In S23, it is determined whether the number of times Noc the overcurrent flag is output has reached a count threshold value Nth.

[0058] If the result is YES in S23, similarly to the first embodiment, the judgment unit 45 judges in S14 that an inverter overcurrent abnormality has occurred, and in S15 stops the supply of current to the inverter 60. If the result is NO in S23 and it is determined in S16 that the monitoring period has elapsed, the judgment unit 45 judges in S17 that the inverter 60 is provisionally normal.

[0059] In an initial check of the monitoring circuit 40 of the second embodiment, the amplifiers 501, 502, and 503 intermittently output a diagnostic monitoring voltage of, for example, 5 V multiple times equal to or greater than the count threshold Nth with respect to a comparator reference value Vref of 4 V. In S03 of the flowchart in Fig. 8, when the number of outputs of the diagnostic monitoring voltage reaches the count threshold Nth, the abnormality determination condition is established.

[0060] The second embodiment is applicable except for cases where the upper arm element or the lower arm element is operated under the condition of always being ON (i.e., the duty ratio is 100% or 0%) and the overcurrent flag is output continuously for multiple periods when stuck ON. Assuming that the duty ratio is greater than 0% and less than 100%, in the second embodiment, if the switching period Tsw is constant, the count proceeds at a constant speed regardless of the duty ratio, and the timing at which the output count Noc reaches the threshold value Nth is constant. Therefore, if the time during which the through current flows is relatively short, it is possible to detect an inverter overcurrent abnormality earlier than in the first embodiment.

[0061] Third embodiment With reference to Fig. 12, inverter overcurrent abnormality detection according to the third embodiment will be described. The third embodiment is a combination of the first and second embodiments. The monitoring circuit 40 includes two types of integrating counters 43, one for integrating the accumulated output time of the overcurrent flag and the other for integrating the number of outputs, and counts both the accumulated output time ΣToc and the number of outputs Noc in parallel.

[0062] The judgment unit 45 judges that an inverter overcurrent abnormality has occurred when the cumulative output time ΣToc of the overcurrent flag in at least one phase reaches a time threshold value Tth, or when the number of outputs Noc of the overcurrent flag reaches a number threshold value Nth, whichever occurs earlier.

[0063] The flow chart of the third embodiment includes S32 and S33 instead of S12 and S13 in Fig. 9. In S32, two types of integrating counters 43 count the cumulative output time ΣToc and the number of outputs Noc of the overcurrent flag for each phase or for the total of the three phases, and the values ​​are held in the corresponding latch circuits 44. In S33, it is determined whether the cumulative output time ΣToc of the overcurrent flag has reached a time threshold Tth, or whether the number of outputs Noc of the overcurrent flag has reached a number threshold Nth. The other steps are similar to those shown in Figs. 9 and 11.

[0064] In the third embodiment, the determination unit 45 determines that an inverter overcurrent abnormality has occurred when either the accumulated output time ΣToc or the number of outputs Noc reaches a threshold value earlier, depending on the duty ratio, etc. This allows the inverter overcurrent abnormality to be detected earlier and the power supply to the inverter 60 to be stopped, thereby further improving reliability.

[0065] (Other embodiments) (a) The load to which the inverter 60 supplies power is not limited to the three-phase motor 80, but may be a multi-phase motor other than three phases, an actuator other than a motor, etc. The number of switching elements in the upper and lower arms of the inverter is not limited to three, but may be one or more.

[0066] (b) The inverter switching elements 61-66 may be composed of other semiconductor switching elements such as FETs other than MOSFETs, bipolar transistors, etc.

[0067] (c) The shunt resistors 67 , 68 , 69 may be provided on the high potential sides of the upper arm elements 61 , 62 , 63 instead of on the low potential sides of the lower arm elements 64 , 65 , 66 .

[0068] (d) The current-monitoring voltage characteristic of the monitoring voltage output device is not limited to the linear characteristic shown in Fig. 4. As long as the monitoring voltage has a positive correlation with the current and the monitoring voltage corresponding to the current in the overcurrent abnormality region can be identified as an abnormal value, the current-monitoring voltage characteristic may be, for example, a stepped or curved characteristic.

[0069] The present invention is not limited to the above-described embodiment, and can be embodied in various forms without departing from the spirit of the present invention.

[0070] The control unit and the method described in the present disclosure may be realized by a special-purpose computer provided by configuring a processor and a memory programmed to execute one or more functions embodied in a computer program. Alternatively, the control unit and the method described in the present disclosure may be realized by a special-purpose computer provided by configuring a processor with one or more dedicated hardware logic circuits. Alternatively, the control unit and the method described in the present disclosure may be realized by one or more special-purpose computers configured by a combination of a processor and a memory programmed to execute one or more functions and a processor configured with one or more hardware logic circuits. In addition, the computer program may be stored in a computer-readable non-transitory tangible recording medium as instructions executed by a computer. [Explanation of symbols]

[0071] 20. Anomaly detection device, 30... Control unit, 40 (40A, 40B) Monitoring circuit, 411, 412, 413... Comparator, 45...judgment section, 501, 502, 503... Amplifier (monitoring voltage output device), 60... Inverter, 61, 62, 63: upper arm elements (switching elements), 64, 65, 66: Lower arm element (switching element), 67, 68, 69... Shunt resistors.

Claims

1. a control unit (30) that performs complementary switching operations on upper arm elements (61, 62, 63) and lower arm elements (64, 65, 66) of a plurality of phases in an inverter (60) in which the upper arm elements and the lower arm elements (64, 65, 66) are bridge-connected; a plurality of monitor voltage output devices (501, 502, 503) for outputting monitor voltages (Vau, Vav, Vaw) having a positive correlation with a current flowing through a shunt resistor (67, 68, 69) connected in series with the upper arm element and the lower arm element of each phase; a monitoring circuit (40) for monitoring an inverter overcurrent abnormality, which is an abnormality in which a through current flows due to the upper arm element or the lower arm element of one or more phases being stuck ON, based on the monitored voltage; Equipped with the monitor voltage output unit holds the output state of the monitor voltage for a predetermined hold time from the input of a trigger signal, The monitoring circuit includes: a comparator (411, 412, 413) that outputs an overcurrent flag of a pulse signal when the monitored voltage is greater than a reference value (Vref); a determination unit (45) that, when a cumulative output time (ΣToc) of the overcurrent flag of at least one phase reaches a time threshold (Tth), determines that an overcurrent abnormality has occurred in the inverter and stops energization of the inverter; An abnormality detection device having the above configuration.

2. a control unit (30) that performs complementary switching operations on upper arm elements (61, 62, 63) and lower arm elements (64, 65, 66) of a plurality of phases in an inverter (60) in which the upper arm elements and the lower arm elements (64, 65, 66) are bridge-connected; a plurality of monitor voltage output devices (501, 502, 503) for outputting monitor voltages (Vau, Vav, Vaw) having a positive correlation with a current flowing through a shunt resistor (67, 68, 69) connected in series with the upper arm element and the lower arm element of each phase; a monitoring circuit (40) for monitoring an inverter overcurrent abnormality, which is an abnormality in which a through current flows due to the upper arm element or the lower arm element of one or more phases being stuck ON, based on the monitored voltage; Equipped with the monitor voltage output unit holds the output state of the monitor voltage for a predetermined hold time from the input of a trigger signal, The monitoring circuit includes: a comparator (411, 412, 413) that outputs an overcurrent flag of a pulse signal when the monitored voltage is greater than a reference value (Vref); a determination unit (45) that determines that an inverter overcurrent abnormality has occurred when a cumulative output time (ΣToc) of the overcurrent flag of at least one phase reaches a time threshold (Tth) or when a number of times the overcurrent flag is output (Noc) reaches a number threshold (Nth), whichever occurs first, and stops energization of the inverter; An abnormality detection device having the above configuration.

3. The control unit, in an initial check of the monitoring circuit, 3. The abnormality detection device according to claim 1, wherein when the monitoring voltage output device outputs a voltage greater than the reference value to the comparator as the monitoring voltage for diagnosis, the comparator outputs the overcurrent flag, and when the cumulative output time or the number of outputs of the monitoring voltage for diagnosis satisfies an abnormality determination condition, the determination unit determines that an inverter overcurrent abnormality has occurred.

4. The abnormality detection device according to any one of claims 1 to 3, wherein the control unit or the monitoring circuit is capable of identifying a phase in which the upper arm element or the lower arm element is stuck ON based on the output of the comparator for each phase.

Images