JP6805094B2 - Internal combustion engine cooling system - Google Patents

Description

The present invention relates to a cooling device for an internal combustion engine.

A technique for learning a reference position of a valve for adjusting the flow of exhaust gas or intake air of an internal combustion engine is known (see Patent Document 1).

JP-A-2015-132204

For example, in some cooling devices of an internal combustion engine, a rotary valve is arranged on a circulation path that passes through the internal combustion engine and circulates a refrigerant. The rotary valve can adjust the flow rate of the refrigerant discharged from the rotary valve to a predetermined position according to the rotation angle position of the rotor. The rotation of the rotor is controlled by the motor, and the drive of the motor is controlled based on the duty ratio of the drive signal applied to the motor.

The rotational resistance of the rotor varies with age. For example, if the rotation resistance of the rotor decreases due to aging, the rotor may rotate too much beyond the target angle even when the command value of the duty ratio of the applied drive signal is the same. In addition, if the rotation resistance of the rotor increases due to aging, the rotation speed of the rotor decreases and the responsiveness decreases, and it may take time for the rotor to reach the target angle, resulting in an increase in power consumption. There is. In order to cope with such aging, it is conceivable to learn the duty ratio of the drive signal required to rotate the rotor as a learning value and control the command value of the duty ratio based on this learning value. ..

However, the rotational resistance of the rotor differs not only with aging but also with the rotational angle position of the rotor and the pressure from the refrigerant received by the rotor. Therefore, if the frequency of updating the learning value as described above is low, there is a possibility that the command value of the duty ratio cannot be controlled so as to be suitable for the rotation of the rotor.

Therefore, an object of the present invention is to provide a cooling device for an internal combustion engine that ensures the frequency of updating the learning value regarding the command value of the duty ratio of the drive signal applied to the motor that drives the rotor of the rotary valve.

The above-mentioned purpose is a rotary valve provided on a circulation path for circulating a refrigerant so as to pass through an internal combustion engine, having a rotor rotated by a motor, and a sensor for detecting the rotation angle position of the rotor, and the sensor. The duty ratio of the drive signal applied to the motor and the deviation detection unit that detects the angle deviation, which is the deviation between the detection angle of the rotor detected by the rotor and the target angle which is the target value of the rotation angle position of the rotor. A storage unit that stores learning values related to command values, a command value calculation unit that calculates the command value based on the angle deviation and the learning value stored in the storage unit, and the calculated command value. A control unit that drives the motor according to the above, a rotation detection unit that detects that the rotor has started rotating in at least one direction by driving the motor, and a rotation detection unit that detects the start of rotation of the rotor each time. A learning value calculation unit that calculates the learning value based on the command value at the start of rotation, an update unit that updates the learning value stored in the storage unit to the calculated learning value, and an internal combustion engine. It can be achieved by the cooling device of. Every time the start of rotation of the rotor is detected, the learning value is calculated and updated based on the command value at the start of rotation, so that the frequency of updating the learning value is ensured.

The storage unit stores the learning value according to the rotation direction of the rotor, and the command value calculation unit stores the learning value stored in the storage unit according to the rotation direction of the rotor, and the rotation direction of the rotor is based on the learning value stored in the storage unit. Separately, the command value is calculated, the rotation detection unit detects the rotation of the rotor according to the rotation direction of the rotor, and the learning value calculation unit calculates the learning value according to the rotation direction of the rotor, and the update The unit may update the learning value stored in the storage unit for each rotation direction of the rotor to the learning value calculated for each rotation direction of the rotor.

The command value calculation unit includes a proportional term according to the angle deviation, an integral term that increases with the passage of time after the application of the drive signal to the motor is started, and the learning stored in the storage unit. The command value may be calculated based on the value.

The command value calculation unit calculates the command value so that the motor reciprocates when the rotation of the rotor is not detected for a predetermined period or more after the application of the drive signal to the motor is started. You may.

The command value calculation unit may calculate the integration term by multiplying the integration value obtained by time-integrating a predetermined fixed value by a predetermined integration gain.

The command value calculation unit may calculate the command value by reducing the learning value so that the learning value corresponds to the dynamic friction force of the rotor.

When the rotor rotated by applying the drive signal to the motor rotates beyond the target angle by a predetermined angle or more, the learning value is corrected to be reduced and stored in the storage unit. A correction update unit that updates the learning value to the corrected learning value may be provided.

The learning value calculation unit may calculate the learning value based on the command value at the rotation start time acquired this time and the command value at the rotation start time acquired last time.

The renewal unit is within a predetermined rotation angle range of the rotor included in the circulation path, connected to the rotary valve, and the opening degree of the flow path changed according to the rotation angle position of the rotor is equal to or less than a predetermined value. If the detection angle belongs, the learning value may be updated, and if the detection angle does not belong within the predetermined rotation angle range, the learning value may not be updated.

The updating unit does not have to update the learning value when the engine rotation speed of the internal combustion engine is equal to or less than a predetermined value, and does not update the learning value when the engine rotation speed exceeds the predetermined value. ..

The rotor may be made of synthetic resin.

According to the present invention, it is possible to provide a cooling device for an internal combustion engine that ensures the frequency of updating the learning value regarding the command value of the duty ratio of the drive signal applied to the motor that drives the rotor of the rotary valve.

FIG. 1 is an explanatory diagram of an engine. FIG. 2 is an explanatory diagram of the cooling device. FIG. 3 is an explanatory view of the rotary valve. FIG. 4 is an explanatory diagram of the rotor. FIG. 5A is a perspective view showing only the rotor and the rotating shaft, and FIG. 5B is a view of the housing as viewed from the introduction port side. FIG. 6 is a graph showing the open / closed state of the radiator flow path, the heater core flow path, and the ATF cooler flow path according to the rotation angle of the rotor. 7A to 7E are schematic cross-sectional views showing an open / closed state of the seal member according to the angle of the rotor. FIG. 8 is a flowchart showing an example of control of the rotary valve executed by the ECU. FIG. 9A is a diagram showing a map for calculating the proportional term, and FIG. 9B is a diagram showing a map for calculating the proportional term in the foreign matter exclusion operation. FIG. 10 is a timing chart when the learning value is updated. FIG. 11 is a timing chart when the rotor is over-rotated. FIG. 12 is a timing chart when the foreign matter exclusion operation is executed.

FIG. 1 is an explanatory diagram of the engine 1. The engine 1 is a so-called V-type engine in which a pair of banks 2L and 2R are arranged around a crankshaft by tilting each other by an appropriate bank angle. Each of the banks 2L and 2R is provided with three cylinders 3. The cylinder 3 is provided with a fuel injection valve 4 for injecting fuel into the cylinder 3. An intake passage 5L and an exhaust passage 7L are connected to the bank 2L, and an intake passage 5R and an exhaust passage 7R are connected to the bank 2R, respectively. The intake passages 5L and 5R are independent of each other, and the exhaust passages 7L and 7R are also independent of each other at least on their respective upstream sides. Throttle valves 6L and 6R for adjusting the intake air amount are provided in the intake passages 5L and 5R, respectively. The engine 1 is an example of an internal combustion engine mounted on a vehicle or the like.

Further, FIG. 1 shows an ECU (Electronic Control Unit) 90. The ECU 90 includes a CPU (Central Processing Unit), a RAM (Random Access Memory), a ROM (Read Only Memory), a memory, and the like. The ECU 90 controls the engine 1 by executing a program stored in a ROM or a memory. The angles of the throttle valves 6L and 6R and the operation of the fuel injection valve 4 and the like are controlled by the ECU 90. Further, the ECU 90 is electrically connected with an accelerator opening sensor S1 for detecting the accelerator opening, a crank angle sensor S2 for detecting the crank angle, and a temperature sensor S3 for detecting the temperature of the refrigerant described later. Has been done.

Next, the cooling device 10 for cooling the engine 1 will be described. FIG. 2 is an explanatory diagram of the cooling device 10. In the cooling device 10, the refrigerant flows through the block water jackets 22L and 22R and the head water jackets 24L and 24R formed on the engine 1, so that heat exchange is performed between the engine 1 and the refrigerant to cool the engine 1. To. A block water jacket 22L and a head water jacket 24L through which the refrigerant flows are formed in each of the cylinder block and the cylinder head of the bank 2L of the engine 1. Similarly, the block water jacket 22R and the head water jacket 24R are formed on the cylinder block and the cylinder head of the bank 2R, respectively. The refrigerant is, for example, LLC (Long Life Coolant).

The cooling device 10 includes a radiator 31, a reserve tank 32, a heater core 33, an ATF (Automatic Transmission Fluid) cooler 35, throttle water jackets 36L and 36R, water pumps 37 and 38, and an inlet 39. The water pump 37 is a mechanical pump that conveys the refrigerant by the rotational power of the engine 1. The refrigerant conveyed from the water pump 37 is supplied in the order of block water jackets 22L and 22R, headwater jackets 24L and 24R, respectively, from the merging flow path 17 via branch flow paths 18L and 18R branched to the left and right. The refrigerant discharged from the headwater jackets 24L and 24R is supplied to the rotary valve 40 via the merging flow path 14 in which the branch flow paths 18L and 18R merge with each other. The radiator flow path 11, the heater core flow path 13, and the ATF cooler flow path 15 are connected to the rotary valve 40, and the radiator flow path 11, the heater core flow path 11, and the heater core flow path are connected according to the rotation angle position of the rotor described later in the rotary valve 40. 13 and each opening degree of the ATF cooler flow path 15 are changed. The drive of the rotary valve 40 is controlled by the ECU 90. Details will be described later.

The radiator flow path 11 is connected between the rotary valve 40 and the inlet 39, and the radiator 31 is provided in the middle. In the radiator 31, heat exchange is performed between the refrigerant and the outside air, and heat dissipation of the refrigerant is promoted. The refrigerant discharged from the radiator 31 is supplied to the inlet 39 via the radiator flow path 11.

A reserve tank flow path 12 is provided, which is branched at a portion upstream of the radiator 31 of the radiator flow path 11 and rejoins the radiator flow path 11 at a portion downstream of the radiator 31 of the radiator flow path 11. A reserve tank 32 is provided in the middle of the reserve tank flow path 12. The reserve tank 32 is provided to absorb a change in volume due to a change in the temperature or pressure of the refrigerant.

The heater core flow path 13 is connected between the rotary valve 40 and the inlet 39, and the heater core 33 is provided in the middle. In the heater core 33, heat exchange is performed between the refrigerant and the air for heating the inside of the vehicle.

A throttle flow path 16 is provided which branches at a portion upstream of the heater core 33 of the heater core flow path 13 and rejoins the heater core flow path 13 at a portion downstream of the heater core 33 of the heater core flow path 13. Throttle water jackets 36R and 36L are provided in the middle of the throttle flow path 16 in order from the upstream side. The throttle water jacket 36R is provided inside the throttle valve 6R, and heat exchange is performed between the refrigerant and the throttle valve 6R in the throttle water jacket 36R to cool the throttle valve 6R. Similarly, the throttle water jacket 36L is provided inside the throttle valve 6L, and heat exchange is performed between the refrigerant and the throttle valve 6L in the throttle water jacket 36L to cool the throttle valve 6R.

A water pump 38 is provided on the downstream side of the point where the heater core flow path 13 merges with the throttle flow path 16. The water pump 38 is an electric pump controlled by the ECU 90, and is provided to assist the transfer of the refrigerant passing through the heater core 33 and the throttle water jackets 36R and 36L. Therefore, when the refrigerant can be sufficiently circulated to the heater core 33 and the throttle water jackets 36R and 36L only by the water pump 37, or when the throttle water jackets 36R and 36L are not provided and the water pump 37 alone is used to reach the heater core 33. The water pump 38 may not be provided if sufficient refrigerant can be circulated.

The ATF cooler flow path 15 is connected between the rotary valve 40 and a portion of the heater core flow path 13 on the downstream side of the water pump 38, and an ATF cooler 35 is provided in the middle. In the ATF cooler 35, heat is exchanged between the ATF used for operating, lubricating, and cooling the automatic transmission fluid and the refrigerant to cool the ATF.

As described above, the refrigerant after passing through the radiator 31, the reserve tank 32, the heater core 33, and the ATF cooler 35 is supplied to the inlet 39, and the refrigerant supplied to the inlet 39 passes through the merging flow path 17 by the water pump 37. It is again supplied to the block water jackets 22L and 22R of the engine 1 via the engine 1.

Next, the rotary valve 40 will be described. FIG. 3 is an explanatory view of the rotary valve 40. The rotary valve 40 includes a housing 41, a rotor 51 rotatably housed in the housing 41, and a drive mechanism 70 for rotationally driving the rotor 51. The housing 41 and rotor 51 are made of synthetic resin.

The housing 41 has a bottom portion formed on the upper end side in FIG. 3 which will be described later, and the lower end side has a substantially tubular shape which is open. The drive mechanism 70 is fixed to the upper end side of the housing 41. An introduction port 45 is formed at the lower end of the housing 41, and the refrigerant that has passed through the headwater jackets 24L and 24R of the engine 1 is supplied into the housing 41 from the introduction port 45. The housing 41 has an upper portion 41a and a lower portion 41b provided so as to be aligned with the axial direction AD of the rotating shaft 59, which will be described later. The upper portion 41a has a substantially cylindrical shape curved so that the central portion in the axial direction AD projects outward in the radial direction. Similarly, the lower portion 41b also has a substantially cylindrical shape in which the central portion in the axial direction AD is curved so as to project outward in the radial direction. The upper portion 41a is formed longer than the lower portion 41b in the axial direction AD. The drive mechanism 70 is provided on the upper stage portion 41a side. In other words, the upper portion 41a is located on the side away from the introduction port 45, and the lower portion 41b is located on the side of the introduction port 45.

As shown in FIG. 3, the upper portion 41a is provided with a discharge portion 431 extending radially outward. Similarly, the lower portion 41b is provided with discharge portions 433 and 435 extending radially outward. The angular spacing around the rotating shaft 59 of the discharging portions 431 and 433 is approximately 90 degrees, and the angular spacing of the discharging portions 433 and 435 around the rotating shaft 59 is also approximately 90 degrees, but is not limited thereto. The discharge units 431, 433, and 435 are opened and closed according to the rotation angle position of the rotor 51. Therefore, the flow rate of the refrigerant introduced into the housing 41 through the introduction port 45 is controlled to pass through the discharge units 431, 433, and 435 according to the rotation angle position of the rotor 51. Here, the radiator flow path 11, the heater core flow path 13, and the ATF cooler flow path 15 described above are connected to the discharge units 431, 433, and 435, respectively. Therefore, the flow rate of the refrigerant supplied to the radiator 31, the heater core 33, and the ATF cooler 35 is controlled according to the rotation angle position of the rotor 51. The inner diameter of the discharge unit 431 is larger than that of the discharge units 433 and 435, respectively, and the inner diameter of the radiator flow path 11 is also larger than that of the heater core flow path 13 and the ATF cooler flow path 15.

The drive mechanism 70 includes a case 71, a motor 72, gears 73 to 76, and an angle sensor 78. The case 71 houses the motor 72, the gears 73 to 76, and the angle sensor 78, and is fixed to the upper end side of the housing 41. The rotation angle of the motor 72 is controlled by the ECU 90. The gear 73 is fixed to the rotating shaft of the motor 72. The gears 74 and 75 are rotatably supported within the case 71. The gear 74 meshes with the gear 73, and the gear 75 meshes with the gear 74. The gear 76 meshes with the gear 75, and the rotating shaft 59 is connected to the center of the gear 76. In this way, the power of the motor 72 is transmitted to the rotating shaft 59 via the gears 73 to 76. The rotor 51 rotates integrally with the rotating shaft 59. Therefore, the rotor 51 is rotated by the motor 72.

The angle sensor 78 is a non-contact type sensor that detects the rotation angle position of the gear 76. Specifically, the angle sensor 78 is an element in which the output signal to the ECU 90 changes according to the magnetic flux density that changes depending on the relative position with the magnetic material (not shown) that rotates with the gear 76, and is, for example, a Hall element. .. Here, as described above, the rotary shaft 59 is connected to the gear 76, and the rotary shaft 59 is connected to the rotor 51. Therefore, the ECU 90 can acquire the detection angle of the rotor 51 detected by the angle sensor 78 based on the output signal of the angle sensor 78. Instead of the angle sensor 78, for example, a potentiometer that rotates with the rotation shaft 59 may be adopted. This is because the rotation angle position of the rotor 51 can be detected via the rotation shaft 59 by the potentiometer. Further, instead of the angle sensor 78, a sensor that detects the rotation angle position of the rotor of the motor 72 may be used to detect the rotor 51 to detect the angle of the rotor 51.

Note that FIG. 3 also shows a functional block diagram of the ECU 90. The ECU 90 includes a deviation detection unit 91, a storage unit 92, a command value calculation unit 93, a control unit 94, a rotation detection unit 95, and a learning value calculation unit 96, which are functionally realized by the CPU, ROM, RAM, and memory described above. The rotary valve 40 is controlled by the update unit 97 and the correction update unit 98. These will be described in detail later.

Next, the rotor 51 will be described. FIG. 4 is an explanatory diagram of the rotor 51. The rotor 51 has a substantially tubular shape in which an upper end surface 58 on the upper end side is formed in FIG. 4 and an introduction port 55 is formed on the lower end side. The rotor 51 is housed in the housing 41 so that the upper end surface 58 side is located on the drive mechanism 70 side and the introduction port 55 is on the introduction port 45 side. Therefore, the refrigerant is introduced from the engine 1 into the rotor 51 via the introduction ports 45 and 55. A stopper portion 58s is formed on the upper end surface 58, which will be described in detail later.

The rotor 51 has an upper portion 51a and a lower portion 51b arranged in the axial direction AD. The upper portion 51a has a substantially cylindrical shape curved so that the central portion in the axial direction AD protrudes outward in the radial direction. Similarly, the lower portion 51b also has a substantially cylindrical shape in which the central portion in the axial direction AD is curved so as to project outward in the radial direction. The upper portion 51a is formed longer than the lower portion 51b in the axial direction AD. The upper portion 51a and the lower portion 51b of the rotor 51 are located inside the upper portion 41a and the lower portion 41b of the housing 41, respectively, and have similar shapes. The upper portion 51a and the lower portion 51b are formed with an upper discharge port 531 and a lower discharge port 533 extending in the circumferential direction, respectively. The upper discharge port 531 and the lower discharge port 533 have different lengths in the circumferential direction, positions in the circumferential direction, and widths in the axial direction AD. Specifically, the circumferential length of the upper discharge port 531 is shorter than that of the lower discharge port 533, and the width of the upper discharge port 531 in the axial direction AD is wider than that of the lower discharge port 533.

FIG. 4 shows the seal members 61, 63, and 65. The seal members 61, 63, and 65 are arranged so as to be movable within a predetermined range in the respective axial directions within the discharge portions 431, 433, and 435, respectively, and are urged toward the rotor 51 side. The seal member 61 has a substantially cylindrical tubular portion 611 and a substantially annular flange portion 613 having a diameter larger than that of the tubular portion 611 at the end on the upper stage portion 51a side of the tubular portion 611. The outer diameter of the flange portion 613 is formed to be slightly larger than the width of the upper discharge port 531 in the axial direction AD, and the flange portion 613 is formed so as not to be inserted into the rotor 51 via the upper discharge port 531. ing. Similarly, the seal member 65 also has a substantially cylindrical tubular portion 651 and a substantially annular flange portion 653 having a diameter larger than that of the tubular portion 651 at the end portion of the tubular portion 651 on the lower stage portion 51b side. The outer diameter of the flange portion 653 is formed to be slightly larger than the width of the lower discharge port 533 in the axial direction AD, and the flange portion 653 is formed so as not to be inserted into the rotor 51 via the lower discharge port 533. ing. Similarly, the seal member 63 also has a substantially cylindrical tubular portion 631 and a substantially annular flange portion (not shown) having a diameter slightly larger than that of the tubular portion 631 at the end portion of the tubular portion 631 on the lower stage portion 51b side. There is. The outer diameter of the flange portion is also formed to be larger than the width of the lower discharge port 533 in the axial direction AD, and the flange portion is formed so as not to be inserted into the rotor 51 via the lower discharge port 533. ..

When the rotor 51 rotates, the flange portion 613 of the seal member 61 slides on the outer peripheral side surface of the upper portion 51a, the relative angular position between the seal member 61 and the upper discharge port 531 is changed, and the overlapping area of the two is changed. To. Therefore, the flow rate of the refrigerant discharged from the rotor 51 to the radiator 31 via the seal member 61 and the discharge unit 431 is changed according to the rotation angle position of the rotor 51. Similarly, when the rotor 51 rotates, the flange portion of the seal member 63 and the flange portion 653 of the seal member 65 slide on the outer peripheral side surface of the lower stage portion 51b, and the relative angular position between the lower stage discharge port 533 and the discharge portion 433 is changed. The overlapping area of the two is changed, the relative angular position between the lower discharge port 533 and the discharge portion 435 is changed, and the overlapping area of the two is changed. Therefore, the flow rate of the refrigerant discharged from the rotor 51 to the heater core 33 via the seal member 63 and the discharge unit 433 is changed according to the rotation angle position of the rotor 51, and the seal member 65 and the discharge unit 435 are changed. The flow rate of the refrigerant discharged to the ATF cooler 35 via the ATF cooler 35 is also changed.

A coil-shaped coil spring 61s is wound around the outer circumference of the tubular portion 611. One end of the coil spring 61s is fixed to the inner surface side of the discharge portion 431, and the other end urges the flange portion 613 of the seal member 61 toward the upper portion 51a. Therefore, the flange portion 613 of the seal member 61 is constantly pressed against the upper portion 51a by the urging force of the coil spring 61s. As a result, the flange portion 613 of the seal member 61 is separated from the upper portion 51a, and the refrigerant leaks from this gap to prevent the flow rate of the refrigerant that should be discharged from the discharge portion 431 from decreasing. Similarly, a coil-shaped coil spring 63s is wound around the outer circumference of the tubular portion 631, one end of the coil spring 63s is fixed to the inner surface side of the discharge portion 433, and the other end is the flange portion of the seal member 63 at the lower stage portion 51b. We are urging towards. A coil-shaped coil spring 65s is wound around the outer circumference of the tubular portion 651, one end of the coil spring 65s is fixed to the inner surface side of the discharge portion 435, and the other end urges the flange portion 653 toward the lower portion 51b. ing. Therefore, the flange portion of the seal member 63 and the flange portion 653 of the seal member 65 are separated from the lower portion 51b, and the refrigerant leaks from this gap, and the flow rate of the refrigerant that should be discharged from the discharge portion 433 and the discharge portion 435 increases. The decrease is suppressed.

The pressure from the refrigerant received by the rotary valve 40 mainly depends on the mechanical water pump 37 linked to the rotation of the engine 1. That is, when the engine 1 is rotating at high speed, the pressure from the refrigerant received by the rotary valve 40 also increases. Therefore, the urging force of the coil springs 61s, 63s, and 65s is such that even when the engine 1 is rotating at high speed and the pressure of the refrigerant received by the rotary valve 40 is large, the sealing members 61, 63, and 65 from the rotor 51 It is set large so that it does not separate.

Next, the rotatable range of the rotor 51 will be described. FIG. 5A is a perspective view showing only the rotor 51 and the rotating shaft 59. As described above, the stopper portion 58s has a substantially fan shape centered on the rotation shaft 59 and is formed so as to project from above the upper end surface 58. FIG. 5B is a view of the housing 41 as viewed from the introduction port 45 side. In FIG. 5B, the drive mechanism 70 is omitted. On the inner upper surface 48 facing the side where the introduction port 45 of the housing 41 is formed, a substantially fan-shaped stopper portion 48s centered on the through hole 49 is provided around the through hole 49 through which the rotating shaft 59 penetrates. It is formed protruding from. When the rotor 51 is housed in the housing 41, the stopper portions 48s and 58s face each other in the circumferential direction. Therefore, when the rotor 51 rotates in one direction beyond a predetermined angle range, one side surface of the stopper portion 58s comes into contact with one side surface of the stopper portion 48s, and further rotation of the rotor 51 is restricted. Similarly, when the rotor 51 rotates in the other direction beyond a predetermined angle range, the other side surface of the stopper portion 58s comes into contact with the other side surface of the stopper portion 48s, and further rotation of the rotor 51 is restricted. In this way, the rotatable range of the rotor 51 is regulated. The stopper that regulates the rotation range of the rotor 51 is not limited to such a shape and position. Further, for example, the gear 76 and the case 71 may be provided with stopper portions, respectively, to indirectly regulate the rotation range of the rotor 51.

FIG. 6 is a graph showing the open / closed state of the radiator flow path 11, the heater core flow path 13, and the ATF cooler flow path 15 according to the rotation angle of the rotor 51. FIG. 6 shows an angle αS to an angle (−βS), which is a rotatable angle range of the rotor 51 regulated by the stopper portions 48s and 58s described above. The angle that serves as a reference for the rotation of the rotor 51 is 0 degrees. FIG. 6 shows a case where the rotor 51 rotates counterclockwise when viewed from the upper end surface 58 side of the axial direction AD toward the right side from the angle of 0 degree. In FIG. 6, the rotor 51 rotates clockwise when viewed from the upper end surface 58 side of the axial direction AD as the angle advances from 0 degree to the left side. Further, the angle of the rotor 51 when rotated counterclockwise from an angle of 0 degrees is indicated by a positive value, and the angle of the rotor 51 when rotated clockwise from an angle of 0 degrees is indicated by a negative value. Further, FIG. 6 shows the angle α1 to the angle αS of the rotor 51 when rotating counterclockwise from the angle 0 degree, and the angle (−β1) of the rotor 51 when rotating clockwise from the angle 0 degree. Up to (-βS) is shown. The angle α1 to the angle αS is a positive value, and the angle (−β1) to the angle (−βS) is a negative value. In the present specification, the rotation of the rotor 51 in the counterclockwise direction is referred to as forward rotation, and the rotation in the clockwise direction is referred to as reverse rotation.

7A to 7E are schematic cross-sectional views showing the opened / closed states of the seal members 61, 63, and 65 according to the angle of the rotor 51. The upper part of each of FIGS. 7A to 7E shows the open / closed state of the seal member 61, and the lower part shows the open / closed state of the seal members 63 and 65. FIG. 7A shows the open / closed state in the range from the angle (-βS) to the angle (-β4), FIG. 7B shows the open / closed state in the range from the angle (-β1) to the angle α1, and FIG. 7C shows the open / closed state in the range of the angle α2. The open / closed state in the range of ~ angle α3 is shown, FIG. 7D shows the opened / closed state in the range of the angle α4 to the angle α5, and FIG. 7E shows the open / closed state in the range of the angle α5 to the angle α6.

In the range from the angle (-β1) to the angle α1 so as to sandwich the angle 0 degrees, the seal member 61 is fully closed by the outer peripheral side surface of the upper portion 51a as shown in FIG. 7B, and the seal members 63 and 65 are the lower portions. It is fully closed by the outer peripheral side surface of 51b. That is, the radiator flow path 11, the heater core flow path 13, and the ATF cooler flow path 15 are fully closed, and the refrigerant is not discharged from the rotary valve 40 to these flow paths.

When the rotor 51 rotates forward from the angle α1, the lower discharge port 533 begins to overlap the seal member 63 and the angle of the heater core flow path 13 gradually increases, and at the angle α2, the lower discharge port 533 completely overlaps the seal member 63 and the heater core flows. Road 13 is fully open. Further, in the range from the angle α1 to the angle α2, the seal members 61 and 65 are kept fully closed by the outer peripheral side surfaces of the upper portion 51a and the lower portion 51b, respectively. Therefore, in the range from the angle α1 to the angle α2, only the opening degree of the heater core flow path 13 increases as the rotor 51 rotates forward, and only the opening degree of the heater core flow path 13 decreases as the rotor 51 rotates in the reverse direction.

In the range from the angle α2 to the angle α3, the seal member 63 is fully open as shown in FIG. 7C, and the seal members 61 and 65 are maintained fully closed by the outer peripheral side surfaces of the upper portion 51a and the lower portion 51b, respectively. That is, in the range of the angle α2 to the angle α3, only the heater core flow path 13 is fully open regardless of the rotation angle position of the rotor 51, and the radiator flow path 11 and the ATF cooler flow path 15 are kept fully closed.

When the rotor 51 rotates forward from the angle α3, the lower discharge port 533 begins to overlap the seal member 65 and the angle of the ATF cooler flow path 15 gradually increases, and at the angle α4, the lower discharge port 533 completely overlaps the seal member 65 and ATF. The cooler flow path 15 is fully opened. Further, in the range from the angle α3 to the angle α4, the seal member 61 is kept fully closed by the outer peripheral side surface of the upper portion 51a. Therefore, in the range from the angle α3 to the angle α4, when the heater core flow path 13 is fully open, only the angle of the ATF cooler flow path 15 increases as the rotor 51 rotates forward, and the ATF cooler increases as the rotor 51 rotates in the reverse direction. Only the angle of the flow path 15 is reduced.

In the range from the angle α4 to the angle α5, the seal members 63 and 65 are fully open as shown in FIG. 7D, and the seal member 61 is maintained fully closed by the outer peripheral side surface of the upper portion 51a. That is, in the range of the angle α4 to the angle α5, the heater core flow path 13 and the ATF cooler flow path 15 are fully open regardless of the rotation angle position of the rotor 51, and the radiator flow path 11 is kept fully closed.

When the rotor 51 rotates forward from the angle α5, the upper discharge port 531 begins to overlap the seal member 61 and the angle of the radiator flow path 11 gradually increases, and at the angle α6, the upper discharge port 531 completely overlaps the seal member 61 and the radiator flow. Road 11 is fully open. Further, in the range from the angle α5 to the angle α6, the seal members 63 and 65 are maintained fully open as shown in FIG. 7E. Therefore, in the range from the angle α5 to the angle α6, when the heater core flow path 13 and the ATF cooler flow path 15 are fully open, only the angle of the radiator flow path 11 increases as the rotor 51 rotates forward, and the rotor 51 reverses. As it rotates, only the angle of the radiator flow path 11 decreases. The angle αT is set between the angle α5 and the angle α6, which will be described in detail later.

In the range from angle α6 to angle αS, the upper discharge port 531 completely overlaps the seal member 61, the lower discharge port 533 completely overlaps the seal members 63 and 65, and the radiator flow path 11, the heater core flow path 13, and the ATF. The cooler flow path 15 is maintained at full throttle. As described above, the angle αS is the angle at one end of the rotatable range of the rotor 51 regulated by the stopper portions 58s and 48s described above.

When the rotor 51 rotates in the reverse direction from the angle (-β1), the lower discharge port 533 begins to overlap the seal member 65, the angle of the ATF cooler flow path 15 gradually increases, and the lower discharge port 533 completely increases at the angle (-β2). The ATF cooler flow path 15 overlaps with the seal member 65 and is fully opened. Further, in the range from the angle (−β1) to the angle (−β2), the seal members 61 and 63 are kept fully closed by the outer peripheral side surfaces of the upper portion 51a and the lower portion 51b, respectively. Therefore, in the range from the angle (-β1) to the angle (-β2), the angle of the ATF cooler flow path 15 increases as the rotor 51 rotates in the reverse direction while the radiator flow path 11 and the heater core flow path 13 are fully closed. The angle of the ATF cooler flow path 15 decreases as the rotor 51 rotates forward.

In the range from the angle (−β2) to the angle (−β3), the seal member 65 is fully open, and the seal members 61 and 63 are kept fully closed by the outer peripheral side surfaces of the upper portion 51a and the lower portion 51b, respectively. That is, in the range from the angle (-β2) to the angle (-β3), only the ATF cooler flow path 15 is fully open regardless of the rotation angle position of the rotor 51, and the radiator flow path 11 and the heater core flow path 13 are fully closed. Is maintained at.

When the rotor 51 rotates in the reverse direction from the angle (-β3), the upper discharge port 531 gradually begins to overlap the seal member 61, the angle of the radiator flow path 11 gradually increases, and the upper discharge port 531 seals at the angle (-β4). The radiator flow path 11 completely overlaps with the member 61 and is fully opened. Further, in the range from the angle (−β3) to the angle (−β4), the seal member 63 is kept fully closed by the outer peripheral side surface of the lower portion 51b, and the seal member 65 is maintained fully open. Therefore, in the range from the angle (-β3) to the angle (-β4), the angle of the radiator flow path 11 is increased so that the rotor 51 rotates in the reverse direction while the heater core flow path 13 is fully closed and the ATF cooler flow path 15 is fully open. Only increases, and as the rotor 51 rotates forward, only the angle of the radiator flow path 11 decreases. An angle (-βT) angle is set between the angle (-β3) and the angle (-β4), which will be described in detail later.

In the range from the angle (-β4) to the angle (-βS), the seal member 61 and the seal member 65 are fully open as shown in FIG. 7A, and the seal member 63 is maintained fully closed by the outer peripheral side surface of the lower portion 51b. To. That is, in the range from the angle (-β4) to the angle (-βS), the radiator flow path 11 and the ATF cooler flow path 15 are kept fully open and the heater core flow path 13 is fully closed regardless of the rotation angle position of the rotor 51. Is maintained at. As described above, the angle (−βS) is the angle at the other end of the rotatable range of the rotor 51 regulated by the stopper portions 58s and 48s described above. The rotation angle position of the rotor 51 is controlled so as to reach the target angle of the rotor 51 described later, but the target angle is not set to the angle αS and the angle (−βS), and the angles αS are not set. It is set between less than and less than the angle (-βS).

Next, control of rotation of the rotor 51 will be described. The ECU 90 controls the rotation of the rotor 51 by controlling the command value of the duty ratio of the drive signal applied to the motor 72. Specifically, the above command value is calculated based on the proportional term, the integral term, and the learning value described later. The proportional term corresponds to an angle deviation obtained by subtracting the detection angle of the rotor 51 from the target angle which is the target value of the rotation angle position of the rotor 51. The detection angle is detected by the angle sensor 78 as described above. The integral term will be described in detail later.

The learned value is a value related to the duty ratio of the drive signal. Specifically, the ECU 90 calculates the learning value based on the command value of the duty ratio at the time when the rotor 51 starts rotating after starting the application of the drive signal to the motor 72. The reason for this will be explained.

As described above, the coil springs 61s, 63s, and 65s press the seal members 61, 63, and 65 on the rotor 51, respectively, and these are one of the factors that increase the rotational resistance of the rotor 51. Considering the rotational resistance of the rotor 51, it is conceivable to increase the duty ratio of the drive signal to the motor 72 described above in advance. However, if the duty ratio of the drive signal is too large, the rotor 51 may rotate beyond the target rotation position. Therefore, when rotating the rotor 51, it is conceivable to control the duty ratio so as to gradually increase after starting the application of the drive signal to the motor 72. In this case, if the duty ratio of the drive signal at the time when the application of the drive signal to the motor 72 is started is too small, the period from the start of the application of the drive signal to the actual start of rotation of the rotor 51 is long. There is a possibility of becoming. Therefore, even though the rotor 51 does not start rotating, the period during which the drive signal is applied to the motor 72 becomes long, and there is a possibility that the power consumption increases. As described above, if the command value of the duty ratio of the drive signal to the motor 72 is too large or too small, a problem may occur.

Further, the duty ratio of the drive signal to the motor 72 suitable for rotating the rotor 51 may differ due to aging. For example, due to aging, the flange portion 613 of the seal member 61, the flange portion of the seal member 63, the flange portion 653 of the seal member 65, the upper stage portion 51a, and the lower stage portion 51b may wear and the rotational resistance of the rotor 51 may decrease. There is sex. Further, the urging force of the coil springs 61s, 63s, and 65s decreases due to aging, which may also reduce the rotational resistance of the rotor 51. Further, the operating characteristics of the motor 72 and the rotational characteristics of the gears 73 to 76 and the rotating shaft 59 may also change due to changes over time.

Therefore, in this embodiment, the learning value is calculated and updated based on the command value of the duty ratio at the time when the rotor 51 actually starts to rotate, and the optimum drive for rotating the rotor 51 based on this learning value. The command value of the duty ratio of the signal is calculated.

Further, in this embodiment, the learning value is calculated and updated every time the start of rotation of the rotor 51 is detected under a predetermined condition, and the frequency of updating the newly calculated learning value is ensured. The reason for this is as follows. The duty ratio of the drive signal to the motor 72 suitable for rotating the rotor 51 may vary due to a plurality of factors. The plurality of factors include, for example, the pressure received by the rotor 51 from the refrigerant, the rotation angle position of the rotor 51 at the time when the application of the drive signal to the motor 72 is started, and the like.

For example, when the engine 1 is rotating at high speed as described above, the flow rate of the refrigerant introduced into the rotary valve 40 increases, and the pressure from the refrigerant received by the rotor 51 also increases. On the other hand, when the engine 1 is rotating at a low speed, the flow rate of the refrigerant introduced into the rotary valve 40 also decreases, and the pressure from the refrigerant received by the rotor 51 also decreases. Therefore, it is considered that the rotational resistance of the rotor 51 changes according to the change in pressure from the refrigerant received by the rotor 51, and the duty ratio of the drive signal to the motor 72 suitable for rotating the rotor 51 also differs.

Further, the rotational resistance of the rotor 51 also differs depending on the rotational position of the rotor 51 at the time when the application of the drive signal to the motor 72 is started. For example, when the seal member 61 is fully closed by the outer peripheral side surface of the upper portion 51a, the contact area between the flange portion 613 of the seal member 61 and the outer peripheral side surface of the upper portion 51a is large, and the rotor 51 caused by the seal member 61 The rotational resistance of is large. On the other hand, when the seal member 61 overlaps the upper discharge port 531 and the radiator flow path 11 is fully opened, the contact area between the flange portion 613 of the seal member 61 and the outer peripheral side surface of the upper portion 51a is small, and the seal member 61 The resulting rotation resistance of the rotor 51 is small. Similarly, when the seal member 63 is fully closed by the outer peripheral side surface of the lower stage portion 51b, the contact area between the flange portion of the seal member 63 and the outer peripheral side surface of the lower stage portion 51b is large, and the rotor 51 caused by the seal member 63 The rotational resistance of is large. On the other hand, when the seal member 63 overlaps the lower discharge port 533 and the heater core flow path 13 is fully opened, the contact area between the flange portion of the seal member 63 and the outer peripheral side surface of the lower portion 51b is small, which is caused by the seal member 63. The rotational resistance of the rotor 51 is small. Further, when the seal member 65 is fully closed by the outer peripheral side surface of the lower stage portion 51b, the contact area between the flange portion 653 of the seal member 65 and the outer peripheral side surface of the lower stage portion 51b is large, and the rotor 51 caused by the seal member 65 The rotational resistance of is large. On the other hand, when the seal member 65 overlaps the lower discharge port 533 and the ATF cooler flow path 15 is fully opened, the contact area between the flange portion 653 of the seal member 65 and the outer peripheral side surface of the lower portion 51b is small, and the seal member 65 The rotational resistance of the rotor 51 due to the above is small. As described above, the duty ratio of the drive signal to the motor 72 suitable for rotating the rotor 51 differs depending on the rotation angle position of the rotor 51 at the time when the drive signal is applied to the motor 72.

As described above, the degree of influence of these factors on the rotational resistance of the rotor 51 can always change during the driving of the engine 1. Therefore, from the viewpoint of determining the optimum command value of the duty ratio of the drive signal applied to the motor 72 while the engine 1 is being driven, it is preferable that the learning value is updated frequently.

In particular, in this embodiment, the gears 73 to 76 and the rotor 51 are made of synthetic resin, and are lighter than those made of metal. Further, since the rotor 51 is made of synthetic resin, the surface roughness is smaller and the frictional resistance between the seal members 61, 63, and 65 is smaller than that of a rotor made by, for example, cutting out aluminum. .. Therefore, the rotational resistance of the rotor 51 is easily affected by the pressure received by the rotor 51 from the refrigerant, the actual rotational angle position of the rotor 51, and the like. Therefore, if the frequency of updating the learning value is low, it may not be possible to respond to changes in the flow rate of the refrigerant and changes in the actual rotation angle position of the rotor 51, and the rotor 51 may not be rotated appropriately. Therefore, in this embodiment, the frequency of updating the learning value is ensured.

Next, an example of control of the rotary valve 40 executed by the ECU 90 will be described. FIG. 8 is a flowchart showing an example of control of the rotary valve 40 executed by the ECU 90. This control is repeatedly executed at predetermined time intervals.

First, it is determined whether or not the absolute value of the angle deviation, which is the value obtained by subtracting the detection angle of the rotor 51 from the target angle of the rotor 51, is γ or more (step S1). γ is an angle of a positive value, for example, a predetermined angle in the range of about 0.3 to 0.7 degrees. In the case of an affirmative determination in step S1 above, the motor 72 is driven so that the angle of the rotor 51 reaches the target angle, which will be described in detail later. The target angle of the rotor 51 is calculated by the ECU 90 according to the rotation speed and load of the engine 1 and the heating request in the vehicle. Here, the angle deviation itself takes a positive value or a negative value as described above. When the angle deviation is a positive value, it means that the target angle is located on the right side of the actual angle of the rotor 51 in FIG. When the angle deviation is a negative value, the target angle is located on the left side of the actual angle of the rotor 51 in FIG. If a negative determination is made in step S1, the angle of the rotor 51 is considered to have reached the target angle, and this control ends. The process of step S1 is a process executed by the deviation detection unit 91 that detects an angle deviation that is a deviation between the detection angle of the rotor 51 detected by the angle sensor 78 and the target angle that is the target value of the rotation angle position of the rotor 51. This is an example.

In the case of affirmative determination in step S1, the command value [%] of the duty ratio of the drive signal applied to the motor 72 for rotating the rotor 51 so that the angle of the rotor 51 reaches the target angle is calculated. (Step S3). The command value is calculated for each rotation direction of the rotor 51. Specifically, when the above-mentioned angle deviation is positive, the command value is calculated as a positive value so that the rotor 51 rotates forward, and when the angle deviation is negative, the rotor 51 is commanded to rotate in the reverse direction. The value is also calculated as a negative value. The command value is calculated by the following formula (1).
Command value = learning value x coefficient K + proportional term + integral term ... (1)

Here, the learning value [%] is a learning value stored in the RAM of the ECU 90. This learned value is also stored in the RAM of the ECU 90 for each rotation direction of the rotor 51. Specifically, when the angle deviation is positive, the learning value stored as a positive value is used, and when the angle deviation is negative, the learning value stored as a negative value is used. The learning value is calculated based on the command value at the start of rotation when the rotation of the rotor 51 starts, and is stored in the RAM. The learning value and the coefficient K will be described in detail later. The RAM of the ECU 90 is an example of the storage unit 92 that stores the learning value regarding the command value of the duty ratio of the drive signal applied to the motor 72.

The proportional term [%] is calculated based on a map determined according to the angle deviation obtained by subtracting the detection angle from the target angle. This map is stored in the RAM of the ECU 90. FIG. 9A is a diagram showing a map for calculating the proportional term. The horizontal axis shows the angular deviation, and the vertical axis shows the proportional term. If the angle deviation is positive, the proportional term is calculated as a positive value, and if the angle deviation is negative, the proportional term is calculated as a negative value. In this map, it is stipulated that the absolute value of the proportional term decreases as the absolute value of the angle deviation decreases. The calculation of the proportional term is not limited to the map as described above, and may be calculated by a mathematical formula.

The integration term [%] is a value obtained by multiplying an integral value obtained by temporally integrating a predetermined fixed value by a predetermined integral gain. If the angle deviation is positive, the integrated gain is a positive value, and if the angle deviation is negative, the integrated gain is a negative value. As a result, if the angle deviation is positive, the integration term is also calculated as a positive value, and if the angle deviation is negative, the integration term is also calculated as a negative value. The integrated gain is set to a positive value regardless of the positive or negative value of the angle deviation, and the fixed value integrated over time is set to a positive value if the angle deviation is positive and a negative value if the angle deviation is negative. The values may be the same. The details of the integration term will also be described later. The process of step S3 is an example of the process executed by the command value calculation unit 93 that calculates the command value based on the angle deviation and the learning value stored in the RAM.

Next, guard processing is executed for the calculated command value (step S5). Specifically, the command value is limited as follows by the guard process. If the command value is greater than or equal to the upper limit guard value, the command value is limited to the upper limit guard value. If the command value is less than or equal to the lower limit guard value, the command value is limited to the lower limit guard value. When the calculated command value is between the upper limit guard value and the lower limit guard value, the command value is not limited. Here, the upper limit guard value is a positive value and the lower limit guard value is a negative value. By such guard processing, when the absolute value of the command value is extremely large, the command value is limited to the upper limit guard value or the lower limit guard value. As a result, for example, it is possible to suppress the application of a large load to the gears 73 to 76, and the deterioration of their durability is suppressed.

In the above guard processing, when any of the conditions described below is satisfied, the guard value that reduces and corrects the absolute value of the upper limit guard value and the absolute value of the lower limit guard value as compared with the case where it is not satisfied. Decrease correction is performed. Specifically, when the stopper portions 58s and 48s are close to each other, the guard value reduction correction is executed. When the stopper portions 58s and 48s are close to each other and the command value of the duty ratio is large, the rotor 51 rotates and the stopper portion 58s collides with the stopper portion 48s with a large force, and the collision noise increases and the stopper portion Durability of 58s and 48s may also be reduced. Therefore, in this case, the duty ratio command value is further limited by executing the guard value reduction correction. Specifically, when the magnitude of the difference between the detection angle of the rotor 51 and the angle αS is equal to or less than a predetermined value, it is assumed that the stopper portions 58s and 48s are close to each other, and the upper limit guard value is corrected to a smaller value. Will be done. Further, when the magnitude of the difference between the detection angle of the rotor 51 and the angle (-βS) is equal to or less than a predetermined value, the absolute value of the lower limit guard value is smaller because the stopper portions 58s and 48s are close to each other. It is corrected to the value. Further, even when the rotation speed of the rotor 51 is high, the guard value reduction correction is executed. If the rotation speed of the rotor 51 is high, a large load may be applied to the gears 73 to 76 to reduce the durability, and if the rotation speed is too high, the stopper portion 58s may collide with the stopper portion 48s. is there. Therefore, when the rotation speed of the rotor 51 is equal to or higher than a predetermined value, the guard value reduction correction is executed to limit the rotation speed of the rotor 51 and reduce the durability of the gears 73 to 76 and the stopper portions 58s and 48s. Suppress. Specifically, when the rotation speed of the rotor 51 detected based on the output signal of the angle sensor 78 is equal to or higher than a predetermined value, the guard value reduction correction is executed. Further, the guard value reduction correction is executed even when the temperature of the refrigerant is low. When the temperature of the refrigerant is low, the viscosity of the lubricating oil of the oil-impregnated metal bearing that supports the rotation of the rotating shaft 59 increases, and the lubricating oil is applied to the contact surface between the outer peripheral surface of the rotating shaft 59 and the inner peripheral surface of the bearing. If the supply is not sufficient, the contact noise between the rotating shaft 59 and the bearing may increase as the motor 72 rotates. In such a case, by executing the guard value reduction correction, the rotation speed of the rotor 51 is reduced, and the increase in contact noise is suppressed. Specifically, when the temperature of the refrigerant detected based on the temperature sensor S3 is lower than a predetermined value, the guard value reduction correction is executed.

Next, the drive signal of the duty ratio at the command value after the guard process is applied to the motor 72 (step S7). As a result, the motor 72 tries to rotate. The process of step S7 is an example of the process executed by the control unit 94 that drives the motor 72 according to the calculated command value. Next, it is determined whether or not the absolute value of the angle deviation satisfies δ or less (step S9). δ is a positive angle smaller than γ described above, and is, for example, a predetermined angle between about 0.1 and 0.4 degrees. The case of affirmative determination in step S9 will be described later.

In the case of a negative determination in step S9, it is determined whether or not the rotation of the rotor 51 is detected (step S11). Specifically, it is determined whether or not it is detected that the rotor 51 has rotated based on the output signal of the angle sensor 78 after the application of the drive signal to the motor 72 is started. Specifically, it is assumed that the rotor 51 has rotated when the angle at which the rotor 51 has rotated within a predetermined period exceeds a predetermined value, that is, when the angle change rate of the rotor 51 exceeds a predetermined value. Detected. The case of a negative determination in step S11 will be described later. The process of step S11 is an example of the process executed by the rotation detection unit 95 that detects that the rotor 51 has started rotating in at least one direction by driving the motor 72. The above-mentioned "by driving the motor 72" excludes the case where the rotor 51 rotates due to the pressure of the refrigerant, the vibration of the engine 1, or the like even though the drive signal is not applied to the motor 72. The purpose is to do.

If the affirmative determination is made in step S11, it is determined whether or not the detection angle of the rotor 51 belongs to the range of the angle (−βT) from the angle αT (step S13). The angle αT is between the angle α5 at which the radiator flow path 11 starts to open and the angle α6 at which the radiator flow path 11 is fully opened, as shown in FIG. The angle (−βT) is from the angle (−β3) at which the radiator flow path 11 starts to open to the angle (−β4) at which the radiator flow path 11 is fully opened.

Here, as described above, the inner diameter of the radiator flow path 11 is also larger than that of the heater core flow path 13 and the ATF cooler flow path 15, so that the flow rate of the refrigerant flowing through each flow path is large when all of these flow paths are fully opened. , The flow rate in the radiator flow path 11 is the largest. Therefore, when the detection angle of the rotor 51 is between the angle αT and the angle αS, the pressure of the refrigerant that tries to rotate the rotor 51 in the forward direction so that the radiator flow path 11 is fully opened increases significantly. Similarly, when the detection angle of the rotor 51 is between the angle (−βT) and the angle (−βS), the pressure of the refrigerant that tries to reversely rotate the rotor 51 so that the radiator flow path 11 is fully opened increases. Therefore, if the detection angle of the rotor 51 does not belong to the range from the angle αT to the angle (−βT), the command value at the start of rotation of the rotor 51 may not be stable. Therefore, in the case of a negative determination in step S13, the learning value is not calculated or updated based on the command value described later, but the processing after step S3 is executed again and the command value is calculated again. As a result, the reliability of the command value at the start of rotation of the rotor 51 is ensured, and the reliability of the learning value calculated and updated based on this command value is also ensured. Further, when there are a plurality of flow paths that are changed according to the rotation angle position of the rotor 51 as in this embodiment, the opening degree of the radiator flow path 11 having the largest inner diameter among the plurality of flow paths is equal to or less than a predetermined value. In that case, the learning value is calculated and updated. The angle (−βT) from the angle αT is an example within the rotation angle range of the rotor 51 in which the opening degree of the radiator flow path 11 changed according to the rotation angle position of the rotor 51 is equal to or less than a predetermined value.

In the case of an affirmative determination in step S13, it is determined whether or not the rotation speed of the engine 1 is 2500 rpm or less (step S15). When the rotation speed of the engine 1 exceeds 2500 rpm, the vibration of the engine 1 increases, and the vibration is easily transmitted to the rotary valve 40. Here, when the vibration transmitted to the rotary valve 40 is large, it is conceivable that the rotor 51 rotates with a small torque as compared with the case where the vibration is small. This is because if the vibration transmitted to the rotary valve 40 is large, the rotor 51 and the sealing members 61, 63, and 65 also vibrate, and the frictional resistance of the rotation of the rotor 51 with respect to the sealing members 61, 63, and 65 is considered to be small. is there. Therefore, even if the conditions other than the vibration transmitted to the rotary valve 40 are the same, when the vibration transmitted to the rotary valve 40 is large, the command value at the start of rotation of the rotor 51 is higher than when the vibration is small. It may become smaller, and the command value at the start of rotation of the rotor 51 may not be stable. Therefore, in the case of a negative determination in step S15, the learning value is not calculated or updated based on the command value described later, and the processing after step S3 is executed again to calculate the command value. In the case of affirmative determination in step S15, the learning value is calculated or updated based on the command value. As a result, the reliability of the command value at the start of rotation of the rotor 51 is ensured, and the reliability of the learning value calculated and updated based on this command value is also ensured. The rotation speed of the engine 1 is determined based on the output signal from the crank angle sensor S2. Further, the rotation speed is not limited to 2500 rpm.

In the process of step S3, which is executed again after the negative determination is made in step S13 or S15, when the angle deviation is reduced during the period from the previous execution of step S3 to the execution of this time step S3. Is calculated so that the absolute value of the proportional term decreases. Further, the absolute value of the integration term is calculated so as to increase with the elapsed time from the start of application of the drive signal. Here, since the proportional gain and the integral gain are set so that the increase in the absolute value of the integral term is larger than the decrease in the absolute value of the proportional term, the calculated command value increases with the passage of time. To do.

If a positive determination is made in step S15, a new learning value is calculated (step S17). Specifically, it is performed as follows. After the application of the drive signal to the motor 72 is started, the command value of the drive signal to the motor 72 at the time when the rotor 51 starts to rotate as described above, that is, at the time when the rotor 51 starts to rotate. The duty ratio of the drive signal is acquired. Next, a new learning value is calculated based on the acquired command value at the start of rotation of the rotor 51. The process of step S17 is an example of the process executed by the learning value calculation unit 96 that calculates the learning value based on the command value at the start of rotation each time the rotation detection unit 95 detects the start of rotation of the rotor 51. ..

Specifically, this learning value is calculated by the following equation (2).
Learning value = previous learning value + (current command value-previous command value) ÷ coefficient M ... (2)
The command value this time is a command value at the start of rotation this time, and is a command value acquired most recently. The previous command value is a command value at the time of the previous rotation start, and is a command value at the time of the rotation start of the rotor 51 acquired one time before the command value at the time of the current rotation start. The previous learning value is a learning value already stored in the RAM of the ECU 90 at the present time, and is a recently updated learning value. The coefficient M is 2 in this embodiment.

As shown in the equation (2), the learning value is calculated as a value obtained by subtracting the previous command value from the current command value, dividing it by a coefficient M of 2, and adding the previous learning value. Here, the value obtained by subtracting the previous command value from the current command value indicates the change between the current command value and the previous command value. The value obtained by dividing this change by 2 is added to the previous learning value. That is, the smoothing process is executed for the change of the command value, and the change smaller than the actual change is reflected in the learning value calculated this time. For example, it is conceivable that the command value this time is significantly different from the previous command value due to the influence of disturbance such as vehicle vibration. In such a case, if the actual change is reflected in the learning value, the influence of the disturbance is greatly reflected in the learning value, and the reliability of the learning value may decrease. Therefore, in this embodiment, the smoothing process is executed for the actual change as described above, and a new learning value is calculated based on this smoothing value. Therefore, the influence of the disturbance on the learning value is suppressed, and the reliability of the learning value is ensured. Although the coefficient M is 2 in this embodiment, it may be a real number exceeding 1 other than this.

The calculation of the learning value is executed for each rotation direction of the rotor 51 as described above. Therefore, the previous learning value, the current command value, and the previous command value are stored in the RAM of the ECU 90 for each rotation direction. For example, when a drive signal is applied to the motor 72 so that the rotor 51 rotates in the forward direction and the current command value at the start of rotation of the rotor 51 in the positive direction is acquired, this current command value and the previously positive value are obtained. The learning value is calculated based on the previous learning value updated at the time of rotation and the acquired previous command value. Similarly, when a drive signal is applied to the motor 72 so that the rotor 51 rotates in the reverse direction and the current command value at the start of rotation in the reverse direction of the rotor 51 is acquired, this current command value and the previous command value are obtained. The learning value is calculated based on the previously updated learning value and the acquired previous command value when the rotation is reversed to.

The learning value newly calculated in this way is updated with respect to the learning value already stored in the RAM of the ECU 90 (step S19). Also in this case, the learning value is updated for each rotation direction of the rotor 51. For example, when a drive signal is applied to the motor 72 so that the rotor 51 rotates in the forward direction, the learning value calculated this time is updated with respect to the previous learning value in the case of the forward rotation. Similarly, when a drive signal is applied to the motor 72 so that the rotor 51 rotates in the reverse direction, the learning value calculated this time is updated with respect to the previous learning value in the case of the forward rotation. As described above, a new learning value is calculated and updated every time the start of rotation of the rotor 51 is detected under a predetermined condition, so that the update frequency of the learning value is ensured. Therefore, as will be described later, the command value is calculated again based on the new learning value, and the command value suitable for the current situation is calculated. The process of step S19 is an example of the process executed by the update unit 97 that updates the learned value stored in the RAM of the ECU 90 to the calculated learned value.

Next, the command value is calculated again based on the equation (1) based on the newly updated learning value (step S3). The above-mentioned equation (1) is described below again.
Command value = learning value x coefficient K + proportional term + integral term ... (1)
If the command value is calculated again based on the new learning value calculated and updated after the start of rotation of the rotor 51 is detected, the integrated value of the integration term is reset to 0% again.

Here, the coefficient K is a coefficient for converting the learning value that reflects the maximum static friction force equivalent calculated in step S17 into a learning value that reflects the dynamic friction force equivalent that is smaller than the maximum static friction force. Specifically, the coefficient K is a value greater than 0 and less than 1. The learning value calculated in step S17 is calculated based on the command value at the start of rotation of the rotor 51 as described above. Here, the command value at the start of rotation of the rotor 51 can be regarded as the command value at the time when the rotational force of the rotor 51 exceeds the maximum static friction force of the rotor 51. Therefore, it can be considered that the learning value newly calculated in step S17 reflects the equivalent of the maximum static friction force. On the other hand, after detecting the start of rotation of the rotor 51, the rotor 51 can continue to rotate as long as the dynamic friction force smaller than the maximum static friction force is exceeded. For example, when the command value is calculated again using the learning value that reflects the maximum static friction force equivalent after detecting the start of rotation of the rotor 51, the recalculated command value is too large and the rotor 51 rotates beyond the target angle. there's a possibility that. In this embodiment, the rotation of the rotor 51 is performed by multiplying the learning value reflecting the maximum static friction force by the coefficient K and converting the learning value to reflect the dynamic friction force equivalent to the maximum static friction force. The rotor 51 is prevented from rotating beyond the target angle after the start is detected.

The coefficient K is set to a different value according to the detection angle of the rotor 51 and the rotation speed of the engine 1 based on a map or a calculation formula acquired in advance in an experiment. This is because, as described above, the contact areas between the seal members 61, 63, and 65 and the rotor 51 are different depending on the angle of the rotor 51, and the maximum static friction force and the dynamic friction force are also different. Further, the coefficient K is set to a smaller value as the rotation speed of the engine 1 is higher. The higher the rotation speed of the engine 1, the easier it is for the vibration of the engine 1 to be transmitted to the rotary valve 40, and the reliability of the learning value calculated in step S17 decreases. Therefore, the coefficient K is set to a small value in order to prevent the command value from becoming too large from the beginning.

As described above, a new learning value is calculated and updated based on the command value at the start of rotation of the rotor 51 under a predetermined condition, and the processes after step S3 are executed again. Here, the integral value of the integral term of the above-mentioned equation (1) will be further described. In general proportional integral control, the value obtained by time-integrating the angle deviation is used as the integral value, but the integral value in the integral term in this embodiment is different from the general integral value, and a predetermined fixed value is used for time. Use the integrated value. The reason for this is as follows.

As described above, since the target angle is set to a different value according to the operating state of the engine 1, the angle deviation also takes a different value according to the target angle. Therefore, in general proportional integration control, when the angle deviation is a positive value and relatively large, the rate of increase of the integrated value with respect to the elapsed time is relatively large, but the angle deviation is a positive value and relatively small. In some cases, the rate of increase of the integrated value is also relatively small. That is, in general proportional integration control, the rate of change of the integrated value over time is not constant and varies greatly depending on the angular deviation. Therefore, the integral term also varies greatly depending on the angular deviation.

If the time change rate of the integration term is not constant as described above, the angle deviation is different and the time change rate of the integration term is also different even if only the target angle is different and other conditions such as the detection angle are the same. It can be very different. Due to this, the command value at the start of rotation of the rotor 51 may also be significantly different. That is, the command value at the start of rotation of the rotor 51 differs depending on the target angle, and the reliability of the command value at the start of rotation of the rotor 51 may be impaired. Therefore, the integral value in the integral term in this embodiment is different from the general integral value, and a value obtained by time-integrating a predetermined fixed value is used. Therefore, the rate of change of the integral term over time is constant. As a result, it is possible to prevent the command value at the start of rotation of the rotor 51 from being significantly different depending on the target angle. Therefore, the reliability of the command value at the start of rotation of the rotor 51 is ensured, and the reliability of the learning value calculated and updated based on this command value is also ensured.

Next, the case of affirmative determination in step S9 described above will be described. If an affirmative determination is made in step S9, it is considered that the angle of the rotor 51 has reached the target angle, and the application of the drive signal is stopped (step S21).

Next, it is determined whether or not the rotor 51 has over-rotated beyond the target angle by a predetermined angle or more (step S23). Specifically, under the condition that the target angle has not changed substantially, the absolute value obtained by subtracting the angle of the rotor 51 at the time of stopping the application from the detection angle after the lapse of a predetermined minute time after the stop of applying the drive signal is absolute. If the value exceeds a predetermined angle, it is determined that over-rotation has occurred. In this case, it can be considered that the angle of inertial rotation of the rotor 51 after the application of the drive signal is stopped is large and the command value becomes too large. The fact that the target angle does not substantially change includes the case where the target angle does not necessarily have to be the same value and fluctuates within a predetermined range that can be regarded as substantially constant. If a negative determination is made in step S23, this control ends.

If the affirmative determination is made in step S23, the learning value stored in the RAM of the ECU 90 is corrected so as to decrease (step S25). Specifically, the learning value already stored in the RAM of the ECU 90 is multiplied by the coefficient L. The coefficient L is a value greater than 0 and less than 1. Next, the reduced-corrected learning value is updated with respect to the learning value already stored in the RAM of the ECU 90 (step S27). In this way, even when over-rotation is detected, the learning value is reduced and corrected and updated, so that the learning value is updated frequently. As a result, over-rotation is suppressed when the application of the next drive signal is stopped, and the angle of the rotor 51 can be accurately controlled to the target angle. If a negative determination is made in step S23, the processes of steps S25 and S27 are not executed. In the processes of steps S25 and S27, when the rotor 51 rotated by applying the drive signal to the motor 72 rotates beyond the target angle by a predetermined angle or more, the learning value is reduced and the ECU 90 is processed. This is an example of the process executed by the correction update unit 98 that updates the learning value stored in the RAM to the corrected learning value. After the execution of step S27, this control ends.

Next, the case of the negative determination in step S11 described above will be described. If a negative determination is made in step S11, it is determined whether or not the rotor 51 cannot rotate (step S31). Specifically, when the command value of the duty ratio of the drive signal applied to the motor 72 is larger than the predetermined value for a predetermined period and the rotation of the rotor 51 is not detected within this period, It is determined that it cannot rotate. For example, it is conceivable that the rotor 51 cannot rotate due to foreign matter such as dust in the refrigerant being caught between the rotor 51 and the housing 41. If a negative determination is made in step S31, the process of step S3 is executed again. That is, if the rotation of the rotor 51 is not detected (negative determination in step S11), but it is not determined that the rotor 51 is in a non-rotatable state (negative determination in step S31), the command value is calculated again and the guard process is performed. It is executed and a drive signal is applied (steps S3, S5, S7).

If the affirmative determination is made in step S31, the foreign matter exclusion operation for eliminating the foreign matter is executed (step S33). The foreign matter exclusion operation is an operation in which the rotor 51 is forcibly rotated reciprocating a predetermined number of times to deform or crush the foreign matter and remove the foreign matter. Specifically, the target angle is exceptionally switched to the angle αS or the angle (−βS) over a predetermined period, and the command value is calculated so that the rotor 51 alternately rotates forward or reverse. To.

The command value in the foreign matter elimination operation is calculated by the following formula (3). Command value = proportional term ... (3)
Unlike the above equation (1), the integral term and the learning value are not reflected in the command value in the foreign matter exclusion operation. The foreign matter exclusion operation is an operation in which the rotor 51 is forcibly rotated so that it can return to the normal state in the case of an abnormal state in which the rotor 51 cannot rotate with the normal command value, and the command value is set to the time using the integration term. This is because it is not necessary to increase it with the passage of time and it is not necessary to reflect the learning value.

FIG. 9B is a diagram showing a map for calculating the proportional term in the foreign matter exclusion operation. Similar to the map of FIG. 9A, when the angle deviation is positive, the proportional term is calculated as a positive value, and when the angle deviation is negative, the proportional term is calculated as a negative value. The magnitude of the calculated absolute value of the proportional term is defined as a constant value that is relatively large enough to eliminate foreign matter, regardless of the angular deviation. In addition, the proportional term is set to 0% within the absolute value of the angle deviation within a predetermined range, but the target angle is largely switched to the positive side and the negative side with respect to the detection angle in the foreign matter exclusion operation, and the proportional term is 0%. There is virtually no angle deviation.

Next, after the foreign matter removing operation is completed, it is determined whether or not the rotor 51 has only normally rotated during the foreign matter removing operation (step S35). Specifically, when it is detected that the rotor 51 is rotating between the angle αS and the angle (−βS) based on the angle sensor 78 during at least the latter half of the period during which the foreign matter exclusion operation is being executed, the rotor 51 Is determined to have rotated normally. In the above determination, it is not required that the rotor 51 normally rotates between the angles αS and the angle (−βS) during the entire period during the foreign matter exclusion operation. The reason for this is that the foreign matter cannot be removed and does not rotate normally in the initial period during the execution of the foreign matter removing operation, but the foreign matter may be removed and rotate normally in the latter half period. If an affirmative determination is made in step S35, a normal determination is made (step S37), and this control ends. That is, when the normal determination is made, the processes after step S1 are executed again. If a negative determination is made in step S35, an abnormality determination is made (step S39), an output restriction process is executed (step S41), and this control ends. In the output limiting process, the absolute value of the upper limit guard value and the absolute value of the lower limit guard value are reduced and corrected so that the command value is further limited than the above-mentioned guard value reduction correction when the rotor 51 cannot be returned to normal. .. As a result, it is possible to suppress an increase in power consumption due to a large command value being applied even though the rotor 51 cannot rotate normally. When an abnormality is determined, the ECU 90 may urge the driver of the vehicle to replace or repair the rotary valve 40 by turning on a warning light or the like installed in the vehicle interior. Further, when the foreign matter exclusion operation is executed, the learning value is not calculated or updated.

Next, the control executed by the ECU 90 will be described with reference to the timing chart. FIG. 10 is a timing chart when the learning value is updated. FIG. 10 shows the actual angle, the target angle, the proportional term, the integral term, the learning value, and the command value, which are the actual angles of the rotor 51. FIG. 10 shows an example in which the actual angle is a positive value and the target angle is set to 0 degrees. In this case, the proportional term, the integral term, the learning value, and the command value are calculated as negative values so that the rotor 51 rotates in the reverse direction and the actual angle reaches the target angle.

The target angle is set to 0, the command value is calculated, the guard process is executed (steps S3 and S5), and the application of the drive signal to the motor 72 is started at time t1 (step S7). When the application of the drive signal to the motor 72 is started, the rotor 51 starts to rotate in the reverse direction. Since the angular deviation gradually decreases as the rotor 51 begins to rotate in the reverse direction, the absolute value of the proportional term gradually decreases, but at time t2, the absolute value of the integral term begins to increase, which causes the rotation of the rotor 51. The speed gradually increases, the angle change rate of the rotor 51 becomes equal to or higher than a predetermined value at time t3, and the start of rotation of the rotor 51 is detected (affirmative determination in step S11). In the period from time t1 to time t3, a negative judgment is made in steps S3, S5, S7, and S9, a negative judgment is made in S11, and a negative judgment is made in S31. These processes are repeatedly executed, and the absolute value of the command value is gradually increased. Is calculated to increase to. The reason why the value of the integration term is maintained at 0% in the period from time t1 to time t2 is that the rotor 51 starts to rotate immediately after the start of application of the drive signal to the motor 72 and the rotor 51 rotates too much. This is to suppress this, but it is not limited to this. Further, the reason why the rotor 51 is not detected to start rotating before the time t3 is that the rotor 51 starts rotating when the angle change rate of the rotor 51 becomes equal to or more than a predetermined value as described in step S11. This is because it is detected, and the angle change rate of the rotor 51 before the time t3 is smaller than the angle change rate of the rotor 51 at the time t3.

When the start of rotation of the rotor 51 is detected at time t3, a new learning value is calculated and updated (steps S17 and S19), and a new command value is calculated based on the updated learning value (step S3). .. At this time, as described above, the integration term is reset to 0% again to calculate the command value, and at time t4, the drive signal of the duty ratio at this command value is applied to the motor 72 (step S7). Therefore, the angle change rate of the rotor 51 decreases at time t4. When the integration term starts to gradually increase at time t5, the angle change rate of the rotor 51 starts to gradually increase again. When the start of rotation is detected at time t6 (affirmative determination in step S11), a new learning value is calculated and updated again (steps S17 and S19), and a drive signal of the duty ratio at the new command value is sent to the motor 72. Is started (step S7), but in this timing chart, the absolute value of the angular deviation becomes δ or less immediately after the start of application of the drive signal (affirmative determination in step S9), and the application of the drive signal is stopped. Is shown (step S21). Therefore, the command value of the drive signal applied at the time t8 immediately after the time t7 becomes 0%, and the rotor 51 reaches the target angle and stops.

FIG. 11 is a timing chart when the rotor 51 is over-rotated. FIG. 11 shows a case where the rotor 51 inertially rotates even after the command value of the drive signal becomes 0% at the time t8 of FIG. 10, and the rotor 51 overrotates beyond the target angle. FIG. 11 shows the actual angle, the target angle, and the command value of the rotor 51, but as for the learning value, only the learning value for reverse rotation is shown as in FIG.

If the rotor 51 continues to rotate in the reverse direction due to inertia even after the command value becomes 0% at time t8, overrotation of the rotor 51 is detected at time t9 (step S23). As a result, the learning value is reduced, corrected, and updated (steps S25 and S27). Further, at time t9, when the absolute value of the angle deviation is determined to be γ or more (affirmative determination in step S1), the command value for forward rotation of the rotor 51 is calculated based on the learning value for forward rotation, and guard processing is performed. Is executed (steps S3 and S5), a drive signal of this command value at time t10 is applied to the motor 72 (step S7), and the rotor 51 is controlled to rotate in the forward direction.

FIG. 12 is a timing chart when the foreign matter exclusion operation is executed. Note that FIG. 12 shows only the target angle, the actual angle, and the command value. At time t1a, both the target angle and the actual angle are positive values, the actual angle substantially coincides with the target angle, and no drive signal is applied. From this state, when the target angle is a positive value but is set to a value smaller than the actual angle at time t2a, the absolute value of the command value gradually increases, but the actual angle of the rotor 51 changes due to foreign matter. Absent. If the rotation of the rotor 51 is not detected and it is determined that the rotor 51 cannot rotate even after the time t3a when the absolute value of the command value exceeds the predetermined value (affirmative determination in step S31), a foreign object is formed at the time t4a when a predetermined period has elapsed from the time t3a. Execution of the exclusion operation is started (step S33). When the foreign matter exclusion operation is executed, the command value is also switched according to the target angle to be switched, the rotor 51 rotates alternately between the angles αS and the angle (-βS), and the command value becomes 0% at the time t5a. It is set and the foreign matter elimination operation is stopped. When it is determined that the rotor 51 has rotated normally based on the detection angle of the rotor 51 during the foreign matter exclusion operation and a normal determination is made (affirmative determination in step S35, step S37), at time t6a after a predetermined time from time t5a, The target angle is set again to the target angle that was set before the start of the foreign matter exclusion operation, the command value is calculated again, the guard process is performed, and the drive signal is applied (affirmative judgment in step S1, S3, S5, S7), when the actual angle of the rotor 51 reaches the target angle at time t7a (affirmative determination in step S9), the application of the drive signal is stopped (step S21).

Although the examples of the present invention have been described in detail above, the present invention is not limited to such specific examples, and various modifications and modifications are made within the scope of the gist of the present invention described in the claims. It can be changed.

In the above embodiment, the gears 73 to 76, the rotor 51, the seal members 61, 63, and 65 are made of synthetic resin, but the present invention is not limited to this, and at least one of them may be made of metal.

In the above embodiment, the ATF cooler 35 is connected in parallel to the radiator 31 and the reserve tank 32 between the rotary valve 40 and the inlet 39, but the only device connected in this way is the ATF cooler 35. Not limited to. For example, instead of the ATF cooler 35, or in addition to the ATF cooler 35, at least one of the oil cooler and the EGR cooler may be connected in parallel to the radiator 31 and the reserve tank 32.

In the above embodiment, the learning value is newly calculated and updated when the affirmative determination is made in steps S13 and S15 (steps S17 and S19), but the processing of steps S13 and S15 is not always necessary. For example, in the case where the flow rate of the refrigerant is changed from the opening degree of the flow path that is changed according to the rotation angle position, but the torque required for the rotation of the rotor 51 is not easily affected by the flow rate of the refrigerant. , It is not necessary to execute the process of step S13. Further, even when the engine 1 is rotating at high speed, if the vibration of the engine 1 is difficult to be transmitted to the rotary valve 40, the process of step S15 may not be executed.

In the above embodiment, the learning value is calculated and updated for each rotation direction of the rotor 51, but the learning value may be calculated and updated each time the rotor 51 detects the start of rotation in at least one direction. This is because the frequency of updating the learning value when rotating in at least one direction can be ensured.

1 engine (internal combustion engine)
2L, 2R bank 3 cylinders 10 Cooling device 11 Radiator flow path 13 Heater core flow path 15 ATF cooler flow path 31 Radiator 33 Heater core 35 ATF cooler 37, 38 Water pump 40 Rotary valve 41 Housing 431, 433, 435 Discharge section 45, 55 Introduced Port 48s, 58s Stopper 51 Rotor 51a Upper 51b Lower 531 Upper discharge 533 Lower discharge 59 Rotating shaft 61, 63, 65 Sealing member 611, 631, 651 Cylinder 613, 633 Flange 61s, 63s, 65s Coil Spring 70 Drive mechanism 72 Motor 73-76 Gear 78 Angle sensor 90 ECU
91 Deviation detection unit 92 Storage unit 93 Command value calculation unit 94 Control unit 95 Rotation detection unit 96 Learning value calculation unit 97 Update unit 98 Correction update unit

Claims (11)

A rotary valve provided on a circulation path for circulating a refrigerant so as to pass through an internal combustion engine and having a rotor rotated by a motor and a sensor for detecting the rotation angle position of the rotor.
A deviation detection unit that detects an angle deviation that is a deviation between the detection angle of the rotor detected by the sensor and the target angle that is the target value of the rotation angle position of the rotor.
A storage unit that stores a learning value regarding a command value of a duty ratio of a drive signal applied to the motor, and a storage unit.
A command value calculation unit that calculates the command value based on the angle deviation and the learning value stored in the storage unit.
A control unit that drives the motor according to the calculated command value, and
A rotation detection unit that detects that the rotor has started rotating in at least one direction by driving the motor, and
A learning value calculation unit that calculates the learning value based on the command value at the time of rotation start each time the rotation detection unit detects the start of rotation of the rotor.
An update unit that updates the learning value stored in the storage unit to the calculated learning value, and a cooling device for an internal combustion engine. The storage unit stores the learning value for each rotation direction of the rotor, and stores the learning value.
The command value calculation unit calculates the command value for each rotation direction of the rotor based on the learning value stored in the storage unit for each rotation direction of the rotor.
The rotation detection unit detects the rotation of the rotor according to the rotation direction of the rotor.
The learning value calculation unit calculates the learning value for each rotation direction of the rotor.
The cooling device for an internal combustion engine according to claim 1, wherein the updating unit updates the learning value for each rotation direction of the rotor stored in the storage unit with the learning value calculated for each rotation direction of the rotor. The command value calculation unit includes a proportional term according to the angle deviation, an integral term that increases with the passage of time after the application of the drive signal to the motor is started, and the learning stored in the storage unit. The cooling device for an internal combustion engine according to claim 1 or 2, which calculates the command value based on the value. The command value calculation unit calculates the command value so that the motor reciprocates when the rotation of the rotor is not detected for a predetermined period or more after the application of the drive signal to the motor is started. The cooling device for an internal combustion engine according to any one of claims 1 to 3. The cooling device for an internal combustion engine according to claim 3, wherein the command value calculation unit calculates the integral term by multiplying an integral value obtained by time-integrating a predetermined fixed value by a predetermined integral gain. The cooling device for an internal combustion engine according to any one of claims 1 to 5, wherein the command value calculation unit reduces the learning value so that the learning value corresponds to the dynamic friction force of the rotor, and calculates the command value. .. When the rotor rotated by applying the drive signal to the motor rotates beyond the target angle by a predetermined angle or more, the learning value is corrected to be reduced and stored in the storage unit. The cooling device for an internal combustion engine according to any one of claims 1 to 6, further comprising a correction update unit that updates the learning value to the corrected learning value. The learning value calculation unit calculates the learning value based on the command value at the rotation start time acquired this time and the command value at the rotation start time acquired last time. Cooling device for any internal combustion engine of 7. The renewal unit is within a predetermined rotation angle range of the rotor included in the circulation path, connected to the rotary valve, and the opening degree of the flow path changed according to the rotation angle position of the rotor is equal to or less than a predetermined value. The internal combustion engine according to any one of claims 1 to 8, wherein the learning value is updated when the detection angle belongs, and the learning value is not updated when the detection angle does not belong within the predetermined rotation angle range. Cooling device. The updating unit updates the learning value when the engine rotation speed of the internal combustion engine is equal to or less than a predetermined value, and does not update the learning value when the engine rotation speed exceeds the predetermined value. A cooling device for an internal combustion engine according to any one of 9 to 9. The cooling device for an internal combustion engine according to any one of claims 1 to 10, wherein the rotor is made of synthetic resin.

Images