Power conversion circuit

Abstract

Estimate the temperature of the switching element. [Solution] A power conversion circuit comprising: a first switching element; a second switching element connected in series with the first switching element; a cooling device for cooling the first switching element with cooling water; a temperature sensor for detecting the temperature of the cooling water; and a control device for performing a temperature estimation process to estimate the temperature of the first switching element. In the temperature estimation process, the control device performs the steps of: calculating the loss generated per unit time in the first switching element; and calculating the temperature of the first switching element based on the loss and the temperature of the cooling water.

Description

Technical Field

[0004] , , ,

[0005] , , , , ,

[0003] , ,

[0001] The technology disclosed in this specification relates to a power conversion circuit.

[0002] The power conversion circuit disclosed in Patent Document 1 has a plurality of series circuits of switching elements. By switching each switching element, this power conversion circuit supplies an alternating current to a motor. Also, this power conversion circuit can discharge a smoothing capacitor by turning on two switching elements connected in series. Thus, current flows through each switching element of the power conversion circuit in various operations.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] Generally, a switching element used in a power conversion circuit incorporates a temperature detection element (e.g., a diode for temperature detection). However, since the manufacturing process of a switching element incorporating a temperature detection element is complex, using a switching element incorporating a temperature detection element increases costs. In this specification, a technique is proposed that enables estimation of the temperature of a switching element even when the switching element does not incorporate a temperature detection element.

Means for Solving the Problems

[0005] ]] <![CDATA[ ]]The power conversion circuit disclosed herein includes a first switching element, a second switching element connected in series with the first switching element, a cooling device for cooling the first switching element with cooling water, a temperature sensor for detecting the temperature of the cooling water, and a control device for performing a temperature estimation process to estimate the temperature of the first switching element. In the temperature estimation process, the control device performs the steps of calculating the loss that occurs per unit time in the first switching element, and calculating the temperature of the first switching element based on the loss and the temperature of the cooling water.

[0006] In this power conversion circuit, the control device calculates the temperature of the first switching element based on the loss generated per unit time in the first switching element and the temperature of the cooling water. Since the heat generated by the loss in the first switching element is transferred to the cooling water, the temperature of the first switching element can be estimated based on the loss generated per unit time in the first switching element and the temperature of the cooling water. [Brief explanation of the drawing]

[0007] [Figure 1] Circuit diagram of electrical circuit 10. [Figure 2] Circuit diagram of series switch circuit 34. [Figure 3] Block diagram of the cooling mechanism. [Figure 4] A graph showing the operation of the series switch circuit 34 during heat generation. [Figure 5] A loss estimation table that estimates the loss Pd that occurs per pulse. [Figure 6] A graph showing alternating current. [Figure 7] A loss estimation table for estimating the average loss P. [Figure 8] Circuit diagram of a series switch circuit 34 having a gate threshold detection circuit 60. [Figure 9] Circuit diagram of gate threshold detection circuit 60. [Figure 10] Loss estimation tables are provided for each gate threshold. [Figure 11] Circuit diagram of gate threshold detection circuit 60a. [Figure 12] A graph showing an example of characteristic detection using a gate threshold detection circuit. [Figure 13] A circuit diagram of a series switch circuit 34 having multiple switching elements 35 connected in parallel. [Modes for carrying out the invention]

[0008] The electrical circuit 10 shown in Figure 1 is installed in the vehicle. The electrical circuit 10 includes a battery 12, an inverter 30, a motor 80, and a computing device 90. The battery 12 is the vehicle's main battery and outputs DC power. The motor 80 is a three-phase motor that rotates the vehicle's drive wheels. The inverter 30 converts the DC power supplied from the battery 12 into AC power and supplies it to the motor 80. This causes the motor 80 to rotate the drive wheels, and the vehicle moves. The computing device 90 controls each part of the electrical circuit 10. The computing device 90 also has a memory device 92. Various types of data are stored in the memory device 92.

[0009] Motor 80 is a three-phase motor and has three input terminals.

[0010] The inverter 30 has a high-potential input wire 31, a low-potential input wire 32, and three output wires 33U, 33V, and 33W. The high-potential input wire 31 is connected to the positive terminal of the battery 12. The low-potential input wire 32 is connected to the negative terminal of the battery 12. In Figure 1, the high-potential input wire 31 and the low-potential input wire 32 are directly connected to the battery 12, but they may also be connected to the battery 12 via another circuit (e.g., a DC-DC converter). A smoothing capacitor 38 is connected between the high-potential input wire 31 and the low-potential input wire 32. Each of the output wires 33U, 33V, and 33W is connected to the corresponding input terminal of the motor 80. Current sensors 39U, 39V, and 39W are provided for each output wire 33U, 33V, and 33W. Each current sensor detects the current flowing through the output wires 33U, 33V, and 33W. The current values ​​detected by each current sensor are input to the calculation unit 90.

[0011] The inverter 30 has three series switch circuits 34U, 34V, and 34W connected between the high-potential input wiring 31 and the low-potential input wiring 32. Each of the series switch circuits 34U, 34V, and 34W has an upper switching element 35U and a lower switching element 35L. The upper switching element 35U and the lower switching element 35L are connected in series between the high-potential input wiring 31 and the low-potential input wiring 32. In series switch circuit 34U, the upper switching element 35U is connected between the high-potential input wiring 31 and the output wiring 33U, and the lower switching element 35L is connected between the output wiring 33U and the low-potential input wiring 32. In series switch circuit 34V, the upper switching element 35U is connected between the high-potential input wiring 31 and the output wiring 33V, and the lower switching element 35L is connected between the output wiring 33V and the low-potential input wiring 32. In the series switch circuit 34W, the upper switching element 35U is connected between the high-potential input wiring 31 and the output wiring 33W, and the lower switching element 35L is connected between the output wiring 33W and the low-potential input wiring 32. The inverter 30 has six gate drive circuits 40. Each gate drive circuit 40 is provided for each switching element 35. Each gate drive circuit 40 controls the corresponding switching element 35 according to a command value input from the arithmetic unit 90.

[0012] The series switch circuits 34U, 34V, and 34W have a common configuration. Figure 2 shows the configuration of each series switch circuit 34. As shown in Figure 2, each switching element 35 has a transistor 36 and a freewheeling diode 37. In Figure 3, the transistor 36 is shown as a MOSFET (metal-oxide-semiconductor field effect transistor). The transistor 36 has a drain as a high-potential terminal and a source as a low-potential terminal. However, the transistor 36 may be composed of an IGBT (insulated gate bipolar transistor). In that case, the transistor 36 has a collector as a high-potential terminal and an emitter as a low-potential terminal. In the upper switching element 35U, the drain of transistor 36U is connected to the high-potential input wiring 31, and the source of transistor 36U is connected to the corresponding output wiring 33. The cathode of the freewheeling diode 37U is connected to the drain of transistor 36U, and the anode of the freewheeling diode 37U is connected to the source of transistor 36U. In the lower switching element 35L, the drain of transistor 36L is connected to the corresponding output wiring 33, and the source of transistor 36L is connected to the low-potential input wiring 32. In addition, the cathode of freewheeling diode 37L is connected to the drain of transistor 36L, and the anode of freewheeling diode 37L is connected to the source of transistor 36L.

[0013] As shown in Figures 1 and 2, each gate drive circuit 40 is connected to the gate of the corresponding switching element 35 transistor 36. Each gate drive circuit 40 controls the gate potential of the transistor 36 to be controlled, with reference to the potential of the source of that transistor 36. Hereinafter, the gate drive circuit 40 that controls the upper switching element 35U will be referred to as gate drive circuit 40U, and the gate drive circuit 40 that controls the lower switching element 35L will be referred to as gate drive circuit 40L.

[0014] As shown in FIG. 3, the vehicle is provided with a cooling mechanism for cooling the inverter 30 and the battery 12. The cooling mechanism includes a cooler 50, a coolant flow path 52, a pump 54, and a water temperature sensor 56. Coolant flows in the coolant flow path 52. The pump 54 circulates the coolant in the coolant flow path 52. The cooler 50 is constituted by a radiator or a heat exchanger, and cools the coolant in the coolant flow path 52 by heat exchange. The coolant flow path 52 is arranged so as to pass through the inverter 30 and the battery 12. The coolant flow path 52 is arranged at a position where heat exchange is possible with each switching element 35 of the inverter 30. When the vehicle is running, each switching element 35 of the inverter 30 is cooled by the coolant flowing in the coolant flow path 52. The water temperature sensor 56 detects the coolant temperature Tw in the coolant flow path 52. The value of the coolant temperature Tw detected by the water temperature sensor 56 is input to the arithmetic unit 90. Note that the water temperature sensor 56 may be provided at any position in the coolant flow path 52. For example, the water temperature sensor 56 may be provided at a position having a high correlation with the temperature of each switching element 35.

[0015] When the vehicle is running, the arithmetic unit 90 executes an inverter operation. In the inverter operation, the arithmetic unit 90 switches the gate potential of each switching element between the gate-off potential VL and the gate-on potential VH, thereby switching each switching element 35. By switching each switching element 35, the arithmetic unit 90 converts the current output from the battery 12 into a three-phase alternating current. The three-phase alternating current output from the inverter 30 is supplied to the motor 80 via the output wirings 33U, 33V, and 33W. The arithmetic unit 90 controls the torque and the rotational speed of the motor 80 by controlling the amplitude and the frequency of the three-phase alternating current supplied to the motor 80. By supplying current to the motor 80 in this way, the motor 80 is driven and the vehicle runs. <000>

[0016] While the vehicle is parked, the battery 12 may be cooled by outside air or the like. If the battery 12 is excessively cooled, the charge / discharge performance of the battery 12 deteriorates. When the temperature of the battery 12 becomes lower than a predetermined temperature, the arithmetic unit 90 executes a warm-up operation to raise the temperature of the battery 12. The warm-up operation will be described below.

[0017] During the warm-up operation, the arithmetic unit 90 circulates the coolant in the coolant flow path 52 by operating the pump 54. Also, during the warm-up operation, the arithmetic unit 90 stops the cooler 50.

[0018] Further, during the warm-up operation, the arithmetic unit 90 executes a heat generation operation in at least one of the series switch circuits 34U, 34V, and 34W. If there is a series switch circuit that does not execute the heat generation operation, the transistors 36U and 36L of that series switch circuit are controlled to be off. Also, if there are a plurality of series switch circuits 34 that execute the heat generation operation, those series switch circuits 34 are controlled synchronously. Since the heat generation operations in each series switch circuit 34 are equal, the heat generation operation of one series switch circuit 34 will be described below.

[0019] Figure 4 shows the operation of the series switch circuit 34 during heating. The gate command value Vi1 is the command value that the arithmetic unit 90 inputs to the gate drive circuit 40U. The gate potential Vg1 is the gate potential of transistor 36U. The gate drive circuit 40U changes the gate potential Vg1 between a high potential Vg1H and a low potential Vg1L according to the gate command value Vi1. The high potential Vg1H is higher than the gate threshold Vth of transistor 36U, and the low potential Vg1L is lower than the gate threshold Vth. The gate drive circuit 40U charges the gate of transistor 36U when the gate command value Vi1 is at a high potential Vi1H, and discharges the gate of transistor 36U when the gate command value Vi1 is at a low potential Vi1L. The gate command value Vi2 is the command value that the arithmetic unit 90 inputs to the gate drive circuit 40L. The gate potential Vg2 is the gate potential of transistor 36L. The gate drive circuit 40L changes the gate potential Vg2 between a high potential Vg2H and a low potential Vg2L according to the gate command value Vi2. The high potential Vg2H is higher than the gate threshold Vth of transistor 36L, and the low potential Vg2L is lower than the gate threshold Vth. The gate drive circuit 40L charges the gate of transistor 36L when the gate command value Vi2 is at the high potential Vi2H, and discharges the gate of transistor 36L when the gate command value Vi2 is at the low potential Vi2L. The short-circuit current I is the current flowing through the series switch circuit 34 (i.e., the current flowing from the high-potential input wiring 31 to the low-potential input wiring 32 via the series switch circuit 34). The loss P is the loss that occurs in transistors 36U and 36L.

[0020] As shown in Figure 4, during the heat generation operation, the arithmetic unit 90 maintains the gate command value Vi1 at a high potential Vi1H. Therefore, the gate drive circuit 40U maintains the gate potential Vg1 at a high potential Vg1H. Consequently, the transistor 36U is always on during the heat generation operation. Also, during the heat generation operation, the arithmetic unit 90 changes the gate command value Vi2 between a high potential Vi2H and a low potential Vi2L at a high frequency. During the period when the gate command value Vi2 is at a high potential Vi2H, the gate drive circuit 40L charges the gate of the transistor 36L, causing the gate potential Vg2 to rise. During the period when the gate command value Vi2 is at a low potential Vi2L, the gate drive circuit 40L discharges the gate of the transistor 36L, causing the gate potential Vg2 to fall. Therefore, during the heat generation operation, the gate potential Vg2 changes between a high potential Vg2H and a low potential Vg2L, causing the transistor 36L to repeatedly switch at a high frequency. During the period Ton when transistor 36L is ON, both transistor 36U and transistor 36L are ON, causing a short circuit between the high-potential input wiring 31 and the low-potential input wiring 32. Therefore, a short-circuit current I flows from the high-potential input wiring 31 through transistors 36U and 36L to the low-potential input wiring 32. During the period Toff when transistor 36L is OFF, the short-circuit current I decreases. In this way, a short-circuit current I repeatedly flows through transistors 36U and 36L during the heating operation. By repeatedly switching transistor 36L, it is possible to flow a short-circuit current I while preventing it from becoming excessive. Due to the flow of the short-circuit current I, losses P repeatedly occur in transistors 36U and 36L during the heating operation. As a result, transistors 36U and 36L generate heat.

[0021] The heat generated by transistors 36U and 36L is transferred to the coolant in the coolant passage 52. As described above, coolant circulates in the coolant passage 52 during warm-up operation (i.e., during heat generation operation). During warm-up operation, the cooler 50 does not cool the coolant, so the temperature of the coolant rises. The heated coolant is supplied to the battery 12, causing the temperature of the battery 12 to rise. In this way, the temperature of the battery 12 can be raised by warm-up operation, and the battery's charge and discharge performance can be restored.

[0022] The arithmetic unit 90 estimates the temperature Tj of each switching element 35. The temperature estimation process performed by the arithmetic unit 90 is described below for each embodiment.

[0023] (Example 1) In Example 1, the computing unit 90 estimates the temperature of the switching element 35 of the series switch circuit 34 that is performing a heat-generating operation. First, the relationship between loss and temperature during the heat-generating operation will be explained. Below, the case in which one series switch circuit 34 is performing a heat-generating operation will be explained as an example. Furthermore, the same relationship between loss and temperature holds for switching element 35U and switching element 35L, so below, the switching element 35L will be explained as an example.

[0024] As described above, the heat generated in the switching element 35L during the heating operation is transferred to the coolant in the coolant flow path 52. Therefore, the temperature of the switching element 35L can be calculated from the coolant temperature Tw and the heat flow rate transferred from the switching element 35L to the coolant. The heat flow rate is approximately equal to the heat generated in the switching element 35L (i.e., the loss). Therefore, the temperature Tj (°C) of the switching element 35L satisfies the following equation (1) with respect to the coolant temperature Tw (°C) and the average loss P (W) generated per unit time in the switching element 35L. Tj = Rth·P + Tw ···(Equation 1) In Equation 1, the symbol Rth represents the thermal resistance (°C / W) between the switching element 35L and the cooling water.

[0025] Furthermore, the average loss P generated in the switching element 35L during the heating operation is obtained by multiplying the loss Pd generated per pulse of the gate potential Vg2 by the frequency fp of the gate potential Vg2. That is, in the heating operation, the average loss P (W), the loss Pd (J) generated per pulse, and the frequency fp (Hz) satisfy the relationship given by equation (2) below. P = Pd·fp ···(Equation 2)

[0026] Furthermore, the longer the on-period Ton of transistor 36L, the higher the short-circuit current I becomes, and the greater the loss Pd per pulse. Therefore, the larger the pulse width W (sec) of the gate-on potential (i.e., the portion higher than the threshold Vth2) of the gate potential Vg2, the greater the loss Pd per pulse. Also, since the current-voltage characteristics of transistor 36L change with temperature, the loss Pd per pulse also changes with the coolant temperature Tw. Therefore, as shown in equation (3) below, the loss Pd per pulse can be expressed as a function f of the pulse width W and the coolant temperature Tw. Pd = f(W, Tw) ... (Equation 3)

[0027] From the above formulas (1), (2), and (3), it can be understood that the temperature Tj of the switching element 35L correlates with the thermal resistance Rth, the coolant temperature Tw, the frequency fp, and the pulse width W. In Example 1, the computing unit 90 does not change the frequency fp during the heat generation operation. Also, the thermal resistance Rth is an intrinsic value of the switching element 35L and can be considered to hardly change. Therefore, the temperature Tj can be estimated from the coolant temperature Tw and the pulse width W.

[0028] Next, the temperature estimation process performed by the arithmetic unit 90 in Example 1 will be described. In Example 1, the memory device 92 of the arithmetic unit 90 stores the loss estimation table shown in Figure 5. The loss estimation table is a graph that shows the relationship between pulse width W and the loss Pd that occurs per pulse. In addition, the graph is defined for each coolant temperature Tw in the loss estimation table. The loss estimation table is created based on experiments, simulations, etc., conducted during the design of the inverter 30.

[0029] The arithmetic unit 90 reads the coolant temperature Tw detected by the water temperature sensor 56 during the heat generation operation. The arithmetic unit 90 also determines the pulse width W based on the gate command value Vi2. Next, the arithmetic unit 90 determines the loss Pd that occurs per pulse based on the coolant temperature Tw, pulse width W, and the loss estimation table in Figure 5. Next, the arithmetic unit 90 performs the calculation of formula (2) based on the determined loss Pd to calculate the average loss P that occurs per unit time. The arithmetic unit 90 stores the frequency fp (e.g., design value) as a fixed value and performs the calculation of formula (2) based on the stored frequency fp. Next, the arithmetic unit 90 performs the calculation of formula (1) based on the calculated average loss P and the measured coolant temperature Tw to calculate the temperature Tj of the switching element 35L. The arithmetic unit 90 stores the thermal resistance Rth (e.g., design value) as a fixed value and performs the calculation of formula (1) based on the stored thermal resistance Rth. In this way, the arithmetic unit 90 can calculate the temperature Tj. Note that the calculations in equations (1) and (2) may be performed together. In this case, the step of calculating the average loss P that occurs per unit time and the step of calculating the temperature Tj based on the average loss P and the cooling water temperature Tw are performed simultaneously.

[0030] The arithmetic unit 90 monitors the temperature Tj by repeatedly calculating it during the heat generation operation. The arithmetic unit 90 monitors the temperature Tj of all switching elements 35 during the heat generation operation. If the temperature Tj of any switching element 35 exceeds a reference value during the heat generation operation, the arithmetic unit 90 performs a process to reduce the temperature Tj of that switching element 35. For example, the arithmetic unit 90 can reduce the pulse width W to reduce the losses occurring in the switching element 35. Therefore, excessive temperature rise of the switching element 35 during the heat generation operation is suppressed.

[0031] (Example 2) In Embodiment 2, the arithmetic unit 90 can change the frequency of the gate command value Vi2 during the heating operation. That is, the arithmetic unit 90 can change the frequency fp of the gate potential Vg2. In addition, in the temperature estimation process of the switching element 35L, the arithmetic unit 90 determines the frequency fp based on the gate command value Vi2 input to the gate drive circuit 40L. The arithmetic unit 90 calculates the average loss P by performing the calculation of formula (2) based on the determined frequency fp. The other configurations of Embodiment 2 are the same as those of Embodiment 1.

[0032] In Example 2, the temperature Tj is calculated according to the frequency fp, so the temperature Tj can be appropriately estimated even when the frequency fp changes.

[0033] Furthermore, since the same short-circuit current I as in transistor 36L flows through transistor 36U, the same losses occur in transistor 36U as in transistor 36L. Therefore, by applying the same temperature estimation process as in Examples 1 and 2 described above to transistor 36U, the temperature of transistor 36U can be estimated.

[0034] Furthermore, in the above embodiments 1 and 2, transistor 36L was switched while transistor 36U was kept ON during the heating operation, but transistor 36U may be switched while transistor 36L is kept ON. In this case as well, the short-circuit current I can be appropriately supplied.

[0035] (Example 3) In Example 3, the computing unit 90 estimates the temperature of the switching element 35 while the inverter is operating (i.e., while the vehicle is running).

[0036] In inverter operation, the arithmetic unit 90 generates three-phase AC currents shown in Figure 6 in output wirings 33U, 33V, and 33W by alternately turning on transistors 36U and 36L in each series switch circuit 34U, 34V, and 34W. In Figure 6, the currents flowing through output wirings 33U, 33V, and 33W are shown as Iu, Iv, and Iw, respectively. The arithmetic unit 90 switches each transistor 36 at a predetermined carrier frequency fc. In this embodiment, the carrier frequency fc is a fixed value. The arithmetic unit 90 also controls the amplitude and frequency of currents Iu, Iv, and Iw to target values ​​by changing the on-duty cycle of each transistor 36 based on the currents Iu, Iv, and Iw detected by current sensors 39U, 39V, and 39W.

[0037] Next, the current flowing through each series switch circuit 34 will be explained. Since the operation of each series switch circuit 34 is the same, the series switch circuit 34U will be explained below. The arithmetic unit 90 alternately turns on transistors 36U and 36L of the series switch circuit 34U. When transistor 36U is turned on while current is flowing in the forward direction (i.e., from output wiring 33U towards motor 80) in output wiring 33U, current flows from high-potential input wiring 31 through transistor 36U to output wiring 33U. Also, when transistor 36U is turned off while current is flowing in the forward direction in output wiring 33U, a freewheeling current flows through the freewheeling diode 37L of switching element 35L. If transistor 36L is turned on at this timing, the freewheeling current splits and flows through transistor 36L and freewheeling diode 37L. When transistor 36L is turned on while current is flowing in the reverse direction (from motor 80 to output wiring 33U), current flows from output wiring 33U through transistor 36L to low-potential input wiring 32. Also, when transistor 36L is turned off while current is flowing in the reverse direction to output wiring 33U, a freewheeling current flows through the freewheeling diode 37U of switching element 35U. If transistor 36U is turned on at this timing, the freewheeling current splits and flows through transistor 36U and freewheeling diode 37U.

[0038] Next, we will explain the losses that occur in the switching elements 35U and 35L during the period in which the alternating current Iu flows through the output wiring 33U for one cycle (i.e., the period Tu in Figure 6). Since the types of losses that occur in the switching elements 35U and 35L are the same, we will explain using the switching element 35L as an example below. During the period Tu, various losses occur in the switching element 35L. When the transistor 36L is turned on and current flows through it, a steady-state loss Esat(J) occurs in the transistor 36L. The steady-state loss Esat correlates with the ratio of the period in which the transistor 36L is turned on (i.e., the on-duty cycle). Also, when the transistor 36L switches, a turn-on loss Pon(W) and a turn-off loss Poff(W) occur. When a return current flows through the switching element 35L, a conduction loss Ef(W) due to the return current occurs. When the freewheeling current splits and flows through the freewheeling diode 37L and transistor 36L, the total loss in the freewheeling diode 37L and transistor 36L is the conduction loss Ef. The conduction loss Ef correlates with the ratio of the time that transistor 36U is off (i.e., the off-duty cycle). Also, when the freewheeling current stops, a recovery loss Prr (J) occurs in the freewheeling diode 37L. The turn-on loss Pon, turn-off loss Poff, and recovery loss Prr occur each time transistors 36U and 36L switch. Therefore, the value obtained by multiplying the losses Pon, Poff, and Prr by the carrier frequency fc is the average value (W) of losses Pon, Poff, and Prr per unit time. For this reason, in inverter operation, the average loss P generated per unit time in the switching element 35L satisfies the relationship in equation 4 below. P = (Pon + Poff + Prr) * fc + Esat + Ef ... (Equation 4)

[0039] During inverter operation, the temperature Tj of the switching element 35L satisfies the relationship shown in equation (1) above. From equations (1) and (4) above, it can be understood that the temperature Tj of the switching element 35L correlates with the thermal resistance Rth, coolant temperature Tw, carrier frequency fc, turn-on loss Pon, turn-off loss Poff, recovery loss Prr, steady-state loss Esat, and conduction loss Ef.

[0040] In Example 3, the arithmetic unit 90 does not change the carrier frequency fc. Also, since the on-duty cycle and off-duty cycle of each switching element 35 are controlled according to the target RMS value Ie of the AC current Iu, the losses Esat and Ef correlate with the RMS value Ie of the AC current Iu (i.e., the value obtained by dividing the amplitude Im in Figure 6 by the square root of 2). For this reason, the average loss P that satisfies the above formula (4) correlates with the RMS value Ie. Furthermore, since the current-voltage characteristics of the switching element 35L change with temperature, the average loss P generated in the switching element 35L changes with the coolant temperature Tw. That is, the average loss P correlates with the RMS value Ie and the coolant temperature Tw. Therefore, the temperature Tj can be estimated from the coolant temperature Tw and the RMS value Ie.

[0041] Next, the temperature estimation process of Embodiment 3, executed by the arithmetic unit 90, will be described. In Embodiment 3, the memory device 92 of the arithmetic unit 90 stores the loss estimation table shown in Figure 7. The loss estimation table is a graph that shows the relationship between the effective value Ie and the average loss P. In addition, the loss estimation table has a graph defined for each coolant temperature Tw. The loss estimation table is created based on experiments, simulations, etc., conducted during the design of the inverter 30.

[0042] The calculation unit 90 reads the coolant temperature Tw detected by the water temperature sensor 56 during inverter operation. The calculation unit 90 also determines the effective value Ie of the current Iu based on the measurement value from the current sensor 39U. Next, the calculation unit 90 determines the average loss P based on the coolant temperature Tw, the effective value Ie, and the loss estimation table in Figure 7. Then, the calculation unit 90 calculates the temperature Tj of the switching element 35L by performing the calculation using formula (1) based on the determined average loss P and the measured coolant temperature Tw. The calculation unit 90 stores the thermal resistance Rth (e.g., design value) as a fixed value and performs the calculation using formula (1) based on the stored thermal resistance Rth. In this way, the calculation unit 90 can calculate the temperature Tj.

[0043] The calculation unit 90 monitors the temperature Tj by repeatedly calculating it during inverter operation. The calculation unit 90 monitors the temperature Tj of all six switching elements 35 during inverter operation. If the temperature Tj of any switching element 35 exceeds a reference value during inverter operation, the calculation unit 90 performs a process to reduce the temperature Tj of that switching element 35. For example, the temperature Tj of each switching element 35 can be reduced by changing the on-duty cycle of each switching element 35. This suppresses excessive temperature increases of the switching elements 35.

[0044] (Example 4) In Embodiment 4, the electrical circuit 10 has a gate threshold detection circuit 60 as shown in Figure 8. A gate threshold detection circuit 60 is provided for each transistor 36. The gate threshold detection circuit 60 is connected to the gate of the corresponding transistor 36. The gate threshold detection circuit 60 detects the gate threshold of the corresponding transistor 36.

[0045] Figure 9 shows the gate threshold detection circuit 60 of Embodiment 4. The gate threshold detection circuit 60 includes a power supply wiring 61, a constant current source 62, a DESAT diode 63, a power supply 64, a comparator 65, a comparator 66, a hold circuit 67, and a memory 68. The power supply wiring 61 is wiring to which a potential VCC is applied. The potential VCC is higher than the source potential of the transistor 36. The input terminal of the constant current source 62 is connected to the power supply wiring 61. The output terminal 62a of the constant current source 62 is connected to the anode of the DESAT diode 63. Alternatively, a pull-up resistor may be provided instead of the constant current source 62. The cathode of the DESAT diode 63 is connected to the drain of the transistor 36. The DESAT diode 63 is a diode for detecting the drain potential of the transistor 36. For example, the DESAT diode 63 is used to determine whether or not an overcurrent is flowing through the transistor 36. In this embodiment, the DESAT diode 63 is also used for gate threshold detection. Furthermore, the output terminal 62a of the constant current source 62 is connected to the negative input terminal of the comparator 65. A reference potential Vref is applied to the positive input terminal of the comparator 65 by the power supply 64. The output terminal of the comparator 65 is connected to the hold circuit 67. The comparator 65 inputs the potential obtained by subtracting the potential of the negative input terminal from the potential of the positive input terminal to the hold circuit 67. The positive input terminal of the comparator 66 is connected to the gate of the transistor 36. The negative input terminal of the comparator 66 is connected to the source of the transistor 36. The output terminal of the comparator 66 is connected to the hold circuit 67. The comparator 66 inputs the gate potential of the transistor 36 (i.e., the gate potential with respect to the source potential) to the hold circuit 67. When the output potential of the comparator 65 reverses from a negative potential to a positive potential, the hold circuit 67 holds the output value of the comparator 66 (i.e., the gate potential) at that time. The hold circuit 67 outputs the held potential to the memory 68. Memory 68 stores the potential value output by the hold circuit 67.

[0046] Next, the gate threshold detection operation performed by the gate threshold detection circuit 60 will be described. The gate threshold detection operation is performed when inverter operation and warm-up operation are not being performed. In other words, the gate threshold detection operation is performed when transistor 36 is not being used. In the gate threshold detection operation, the arithmetic unit 90 keeps the transistor 36 connected in series with the transistor 36 whose gate threshold is to be detected ON. For example, when measuring the gate threshold of transistor 36L, transistor 36U connected in series with transistor 36L is kept ON. Also, for example, when measuring the gate threshold of transistor 36U, transistor 36L connected in series with transistor 36U is kept ON. In this state, the following processing is performed on the transistor 36 whose gate threshold is to be detected.

[0047] At the start of the gate threshold detection operation, the gate drive circuit 40 controls the gate potential of transistor 36 to the gate-off potential, and transistor 36 is off. In this state, the collector potential of transistor 36 is very high, so the DESAT diode 63 is off. Therefore, the potential VCC of the power supply wiring 61 is applied to the output terminal 62a of the constant current source 62 via the constant current source 62. As a result, the potential VCC is applied to the negative input terminal of the comparator 65. Since the potential VCC is higher than the reference potential Vref, the comparator 65 outputs a negative potential.

[0048] Next, the arithmetic unit 90 increases the gate command value for the gate drive circuit 40. This causes the gate drive circuit 40 to raise the gate potential of transistor 36. Note that the arithmetic unit 90 may increase the gate command value with short pulses to prevent the generation of a high short-circuit current. When the gate potential of transistor 36 reaches the gate threshold Vth, transistor 36 turns on. Then, the drain potential of transistor 36 drops to approximately the same potential as the source potential. Then, the cathode potential of the DESAT diode 63 drops, the DESAT diode 63 turns on, and current flows from the power supply wiring 61 to the source of transistor 36 via the constant current source 62, the DESAT diode 63, and transistor 36. Also, when the DESAT diode 63 turns on, the potential of the output terminal 62a of the constant current source 62 drops to approximately the same potential as the source potential (more specifically, the potential obtained by adding the forward voltage of the DESAT diode 63 to the source potential). As a result, the potential of the negative input terminal of comparator 65 becomes lower than the reference potential Vref, causing the output potential of comparator 65 to invert from a negative potential to a positive potential. The hold circuit 67 holds the output potential of comparator 66 (i.e., the gate potential) at the time the output potential of comparator 65 inverts from a negative potential to a positive potential as its output value. The held potential is the gate potential when transistor 36 is turned on, and therefore corresponds to the gate threshold Vth. The memory 68 stores the gate threshold Vth held by the hold circuit 67 at the end of the gate threshold detection operation. In this way, the gate threshold Vth is detected by the gate threshold detection operation.

[0049] The gate threshold detection circuit 60 periodically performs gate threshold detection processing. For example, the gate threshold detection circuit 60 can perform gate threshold detection processing when the vehicle starts up. As each gate threshold detection circuit 60 performs gate threshold detection processing, the gate thresholds of each of the six transistors 36 are stored in the memory 68.

[0050] Next, the temperature estimation process of Example 4 will be described. The temperature estimation process of Example 4 differs from Example 1 in that it calculates the average loss P according to the gate threshold, but is otherwise the same as Example 1. As shown in Figure 10, the arithmetic unit 90 of Example 4 stores multiple loss estimation tables provided for each gate threshold Vth. In the temperature estimation process of Example 4, the arithmetic unit 90 reads the gate threshold Vth of the transistor 36 to be temperature estimated from the memory 68. Based on the read gate threshold Vth, the arithmetic unit 90 extracts one loss estimation table from the multiple loss estimation tables. Based on the extracted loss estimation table, the coolant temperature Tw, and the pulse width W, the arithmetic unit 90 identifies the loss Pd that occurs per pulse. After that, the arithmetic unit 90 calculates the temperature Tj of the switching element 35 in the same manner as in Example 1 or 2.

[0051] According to Example 4, the temperature Tj is calculated according to the gate threshold Vth, allowing for a more accurate estimation of the temperature Tj. In other words, in transistor 36, an error (i.e., a deviation from the standard value) occurs in the gate threshold Vth. Furthermore, the gate threshold Vth may change over time due to the use of transistor 36. When the gate threshold Vth deviates from the standard value in this way, variations occur in the characteristics of transistor 36. In contrast, in Example 4, the latest gate threshold data is stored in memory 68 by repeating the gate threshold detection process, and the temperature Tj is calculated based on that gate threshold, thus suppressing the effect of variations in the gate threshold Vth. Therefore, the temperature Tj of transistor 36 can be estimated more accurately.

[0052] The gate threshold detection process of Example 4 may be combined with the temperature estimation process of Example 3. That is, in Example 3, the arithmetic unit 90 may store multiple loss estimation tables, as shown in Figure 7, for each gate threshold Vth. In this case, the arithmetic unit 90 reads the gate threshold Vth of the transistor 36 to be temperature estimated from the memory 68 during the temperature estimation process. Based on the read gate threshold Vth, the arithmetic unit 90 extracts one loss estimation table from the multiple loss estimation tables. Using the extracted loss estimation table, the arithmetic unit 90 calculates the temperature Tj of the transistor 36 in the same manner as in Example 3. With this configuration, the influence of variations in the gate threshold Vth can be suppressed, and the temperature Tj can be accurately estimated.

[0053] (Example 5) The gate threshold detection circuit 60a of Embodiment 5 shown in Figure 11 is a circuit that can be used as a substitute for the gate threshold detection circuit 60 of Embodiment 4. The gate threshold detection circuit 60a includes an ammeter 69, a power supply 64, a comparator 65, a hold circuit 67, and a memory 68. The ammeter 69 detects the current flowing through the source of the transistor 36. The ammeter 69 outputs a larger potential the larger the detected current. The output potential of the ammeter 69 is applied to the positive input terminal of the comparator 65. A reference potential Vref is applied to the negative input terminal of the comparator 65 by the power supply 64. The comparator 66, hold circuit 67, and memory 68 are configured in the same way as in Embodiment 4.

[0054] Next, the gate threshold detection operation performed by the gate threshold detection circuit 60a will be described. In the gate threshold detection operation, the arithmetic unit 90 keeps the transistor 36 connected in series with the transistor 36 whose gate threshold is to be detected ON. In this state, the following processing is performed on the transistor 36 whose gate threshold is to be detected.

[0055] At the start of the gate threshold detection operation, the gate drive circuit 40 controls the gate potential of transistor 36 to the gate-off potential, and transistor 36 is turned off. Therefore, the output potential of the ammeter 69 is lower than the reference potential Vref, and the comparator 65 outputs a negative potential.

[0056] Next, the arithmetic unit 90 increases the gate command value for the gate drive circuit 40. This causes the gate drive circuit 40 to raise the gate potential of transistor 36. Note that the arithmetic unit 90 may increase the gate command value with short pulses to prevent the generation of a high short-circuit current. When the gate potential of transistor 36 reaches the gate threshold Vth, transistor 36 turns on. Then, current flows through transistor 36, and the output potential of the ammeter 69 becomes higher than the reference potential Vref. Therefore, the output potential of comparator 65 reverses from a negative potential to a positive potential. The hold circuit 67 holds the output potential of comparator 66 (i.e., the gate potential) at the time the output potential of comparator 65 reverses from a negative potential to a positive potential as its output value. The held potential is the gate potential when transistor 36 is turned on, and therefore corresponds to the gate threshold Vth. The memory 68 stores the gate threshold Vth held by the hold circuit 67 at the end of the gate threshold detection operation. In this way, the gate threshold Vth is detected by the gate threshold detection operation.

[0057] Thus, in Example 5, gate threshold detection is also possible. Therefore, in Example 5, as in Example 4, the temperature Tj of the switching element 35 can be estimated according to the gate threshold.

[0058] In Examples 4 and 5, the reference potential Vref may be variable. In this case, by measuring the gate threshold Vth multiple times while changing the reference potential Vref, the gate potential and drain current curves can be measured as shown in Figure 12. This allows for more accurate measurement of the characteristics of the switching element 35.

[0059] (Example 6) As shown in Figure 13, in Embodiment 6, each series switch circuit 34 has two switching elements 35U connected in parallel, and also has two switching elements 35L connected in parallel. A gate threshold detection circuit 60 is provided for each gate wiring of the two switching elements 35U, and a gate threshold detection circuit 60 is provided for each gate wiring of the two switching elements 35L. Each gate threshold detection circuit 60 may be the gate threshold detection circuit 60 of Embodiment 4, or the gate threshold detection circuit 60a of Embodiment 5. The gate drive circuit 40U controls the two switching elements 35U synchronously in inverter operation. The gate drive circuit 40L controls the two switching elements 35L synchronously in inverter operation. The gate drive circuit 40U controls the gate potential of the two switching elements 35U independently in gate threshold detection operation. The gate drive circuit 40L controls the gate potential of each switching element 35L independently in gate threshold detection operation. In this way, since the gate potential of each switching element 35 can be controlled independently, the gate thresholds of the two switching elements 35 connected in parallel can be detected.

[0060] (Example 7) In the above embodiment, the thermal resistance Rth was treated as a fixed value. However, the thermal resistance Rth between the switching element 35 and the coolant may change over time. In Embodiment 7, the computing device 90 estimates the temperature of the switching element 35 in accordance with the change in thermal resistance Rth over time.

[0061] In Embodiment 7, the arithmetic unit 90 stores a thermal resistance estimation table that represents the relationship between the vehicle's usage history (e.g., vehicle driving time, negative overcurrent history, warm-up time, etc.) and the thermal resistance Rth. The arithmetic unit 90 also records the vehicle's usage history when it is used. When the temperature estimation process is executed, the arithmetic unit 90 reads the vehicle's usage history and identifies the thermal resistance Rth corresponding to that usage history from the thermal resistance estimation table. The arithmetic unit 90 performs the temperature estimation process based on the identified thermal resistance Rth. In this way, the arithmetic unit 90 corrects the thermal resistance used in the temperature estimation process based on the usage history. With this configuration, the temperature can be estimated more accurately.

[0062] In each of the above embodiments, the arithmetic unit 90 and the gate drive circuit 40 are examples of control devices. Furthermore, if the electrical circuit 10 has gate threshold detection circuits 60 and 60a, then the arithmetic unit 90, the gate drive circuit 40, and the gate threshold detection circuits 60 and 60a are examples of control devices.

[0063] The configurations of power conversion circuits disclosed herein are listed below. (Composition 1) A power conversion circuit, First switching element, A second switching element connected in series with the first switching element, A cooling device that cools the first switching element with cooling water, A temperature sensor for detecting the temperature of the cooling water, A control device that performs a temperature estimation process to estimate the temperature of the first switching element, It has, In the temperature estimation process, the control device, A step of calculating the loss that occurs per unit time in the first switching element, A step of calculating the temperature of the first switching element based on the loss and the temperature of the cooling water. A power conversion circuit that performs this operation. (Configuration 2) The power conversion circuit according to configuration 1, wherein the control device calculates the loss based on the pulse width of the gate-on potential in the gate signal of the first switching element in the step of calculating the loss. (Composition 3) The power conversion circuit according to configuration 2, wherein the control device calculates the loss based on the temperature of the cooling water in the step of calculating the loss. (Composition 4) The power conversion circuit according to configuration 2 or 3, wherein the control device calculates the loss based on the frequency of the gate signal of the first switching element in the step of calculating the loss. (Composition 5) The power conversion circuit is capable of performing a warm-up operation to heat the cooling water by repeatedly switching the first switching element while keeping the second switching element in the ON state. The power conversion circuit according to any one of configurations 1 to 3, wherein the control device performs the temperature estimation process during the warm-up operation. (Composition 6) The power conversion circuit has an output wiring connected to the wiring between the first switching element and the second switching element, The power conversion circuit is capable of performing an inverter operation by alternately turning on the first switching element and the second switching element to supply AC current to the output wiring. The control device executes the temperature estimation process during the inverter operation. The power conversion circuit according to configuration 1, wherein the control device calculates the loss based on the amplitude of the alternating current supplied to the output wiring. (Composition 7) The power conversion circuit according to configuration 6, wherein the control device calculates the loss based on the temperature of the cooling water in the step of calculating the loss. (Composition 8) The control device is capable of performing a gate threshold measurement process to measure the gate threshold of the first switching element, The power conversion circuit according to any one of configurations 1, 2, 3, 6, or 7, wherein the control device calculates the loss based on the gate threshold measured in the gate threshold measurement process in the step of calculating the loss. (Composition 9) The power conversion circuit further includes a DESAT diode whose cathode is connected to the high-potential terminal of the first switching element, The control device, in the gate threshold measurement process, measures the gate threshold based on the potential of the anode of the DESAT diode. The power conversion circuit described in Configuration 8. (Composition 10) The power conversion circuit further includes an ammeter for measuring the current flowing through the first switching element, The control device, in the gate threshold measurement process, measures the gate threshold based on the current detected by the ammeter. The power conversion circuit described in Configuration 8. (Composition 11) The power conversion circuit according to any one of configurations 1, 2, 3, 6, or 7, wherein the first switching element is composed of a plurality of switching elements connected in parallel. (Composition 12) The control device has a memory device that stores the thermal resistance of the first switching element, The control device is capable of performing a process to correct the thermal resistance stored in the memory device, In the step of calculating the temperature of the first switching element, the control device calculates the temperature of the first switching element based on the thermal resistance stored in the storage device. A power conversion circuit as described in any one of items 1, 2, 3, 6, or 7.

[0064] In configuration 2, the loss per unit time in the first switching element is calculated based on the pulse width of the gate-on potential. Since the pulse width of the gate-on potential correlates with the loss per unit time in the first switching element, configuration 2 allows for accurate calculation of the loss per unit time in the first switching element.

[0065] In configuration 3, the loss per unit time at the first switching element is calculated based on the pulse width of the gate-on potential and the temperature of the cooling water. The temperature of the cooling water correlates with the temperature characteristics of the first switching element. Therefore, according to configuration 3, the loss per unit time at the first switching element can be suitably calculated more accurately in accordance with the temperature characteristics of the first switching element.

[0066] In configuration 4, the loss per unit time in the first switching element is calculated based on the pulse width of the gate-on potential and the frequency of the gate signal. Since the frequency of the gate signal correlates with the loss per unit time in the first switching element, configuration 3 allows for a more accurate calculation of the loss per unit time in the first switching element.

[0067] According to configuration 5, the temperature of the first switching element can be estimated during the warm-up operation.

[0068] According to configuration 6, the temperature of the first switching element can be estimated during inverter operation. Furthermore, since the amplitude of the AC current supplied to the output wiring correlates with the loss that occurs per unit time in the first switching element, configuration 6 allows for the accurate calculation of the loss that occurs per unit time in the first switching element.

[0069] According to configuration 7, the loss generated per unit time in the first switching element can be suitably calculated more accurately in accordance with the temperature characteristics of the first switching element.

[0070] According to configurations 8-10, the loss per unit time generated in the first switching element can be suitably calculated more accurately in accordance with the gate threshold of the first switching element.

[0071] According to configuration 11, the temperature of each switching element connected in parallel can be detected.

[0072] According to configuration 12, the temperature of the first switching element can be accurately calculated even if the thermal resistance changes over time.

[0073] Although embodiments have been described in detail above, these are merely illustrative and do not limit the scope of the claims. The technologies described in the claims include various modifications and changes to the specific examples illustrated above. The technical elements described in this specification or drawings exhibit technical usefulness individually or in various combinations, and are not limited to the combinations described in the claims at the time of filing. Furthermore, the technologies illustrated in this specification or drawings achieve multiple objectives simultaneously, and achieving even one of these objectives constitutes technical usefulness. [Explanation of Symbols]

[0074] 30: Inverter, 34: Series switch circuit, 35L: Switching element, 35U: Switching element, 56: Water temperature sensor

Claims

1. A power conversion circuit, First switching element, A second switching element connected in series with the first switching element, A cooling device that cools the first switching element with cooling water, A temperature sensor for detecting the temperature of the cooling water, A control device that performs a temperature estimation process to estimate the temperature of the first switching element, It has, In the temperature estimation process, the control device, A step of calculating the loss that occurs per unit time in the first switching element, A step of calculating the temperature of the first switching element based on the loss and the temperature of the cooling water. A power conversion circuit that performs this operation.

2. The power conversion circuit according to claim 1, wherein the control device calculates the loss based on the pulse width of the gate-on potential in the gate signal of the first switching element in the step of calculating the loss.

3. The power conversion circuit according to claim 2, wherein the control device calculates the loss based on the temperature of the cooling water in the step of calculating the loss.

4. The power conversion circuit according to claim 2 or 3, wherein the control device calculates the loss based on the frequency of the gate signal of the first switching element in the step of calculating the loss.

5. The power conversion circuit is capable of performing a warm-up operation to heat the cooling water by repeatedly switching the first switching element while keeping the second switching element in the ON state. The power conversion circuit according to any one of claims 1 to 3, wherein the control device performs the temperature estimation process during the warm-up operation.

6. The power conversion circuit has an output wiring connected to the wiring between the first switching element and the second switching element, The power conversion circuit is capable of performing an inverter operation by alternately turning on the first switching element and the second switching element to supply AC current to the output wiring. The control device executes the temperature estimation process during the inverter operation. The power conversion circuit according to claim 1, wherein the control device calculates the loss based on the amplitude of the alternating current supplied to the output wiring.

7. The power conversion circuit according to claim 6, wherein the control device calculates the loss based on the temperature of the cooling water in the step of calculating the loss.

8. The control device is capable of performing a gate threshold measurement process to measure the gate threshold of the first switching element, The power conversion circuit according to any one of claims 1, 2, 3, 6, or 7, wherein the control device calculates the loss based on the gate threshold measured in the gate threshold measurement process in the step of calculating the loss.

9. The power conversion circuit further includes a DESAT diode whose cathode is connected to the high-potential terminal of the first switching element. The control device, in the gate threshold measurement process, measures the gate threshold based on the potential of the anode of the DESAT diode. The power conversion circuit according to claim 8.

10. The power conversion circuit further includes an ammeter for measuring the current flowing through the first switching element, The control device, in the gate threshold measurement process, measures the gate threshold based on the current detected by the ammeter. The power conversion circuit according to claim 8.

11. The power conversion circuit according to any one of claims 1, 2, 3, 6, or 7, wherein the first switching element is composed of a plurality of switching elements connected in parallel.

12. The control device has a memory device that stores the thermal resistance of the first switching element, The control device is capable of performing a process to correct the thermal resistance stored in the memory device, In the step of calculating the temperature of the first switching element, the control device calculates the temperature of the first switching element based on the thermal resistance stored in the storage device. A power conversion circuit according to any one of claims 1, 2, 3, 6, or 7.

Images