Rotation angle detection device

Abstract

Provided is a rotation angle detection device capable of determining abnormality of the rotation angle detection device only by detection information of the rotation angle detection device even in a case where the rotation speed is low. The rotation angle detection device aggregates sensor output values of N-1 rotation detection sensors other than a diagnosis target sensor, calculates a value obtained by reversing the sign of the aggregated value as an estimated value of the sensor output value of the diagnosis target sensor, calculates a difference between the sensor output value of the diagnosis target sensor and the estimated value of the sensor output value of the diagnosis target sensor, i.e., an estimated difference, and determines that the diagnosis target sensor is abnormal in a case where the estimated difference is outside a determination range for the estimated difference.

Description

Technical Field

[0001] This application relates to a rotation angle detection device. Background Technology

[0002] In recent years, to reduce the environmental impact of vehicles, the electrification of vehicles using electric motors for propulsion has been gradually advancing. In such electric vehicles, it is necessary to accurately detect the rotation angle of the electric motor to ensure the vehicle's driving performance. Furthermore, when an abnormality is detected in the rotation angle detection device, it is required to immediately detect the fault and transition to a safe state.

[0003] As a fault diagnosis method for rotation angle detection devices, there is, for example, Patent Document 1. Patent Document 1 discloses an anomaly detection method by comparing the motor angle calculated based on the output mode of a Hall sensor that outputs according to the rotational position of the motor rotor with an estimated motor angle calculated based on the reverse voltage of the motor. Existing technical documents Patent documents

[0004] Patent Document 1: Japanese Patent Application Publication No. 2005-335591 Summary of the Invention The technical problem that the invention aims to solve

[0005] However, in the anomaly detection of the rotation angle detection device in Patent Document 1, anomalies cannot be detected until the timing of the Hall sensor's output signal switching. When the motor speed is low, the time until an anomaly is detected becomes quite long. Furthermore, when the motor speed is low, the induced voltage is low, and since the induced voltage cannot accurately calculate the estimated motor angle, the anomaly detection accuracy becomes low. Additionally, since the estimated motor angle is based on the reverse voltage, anomalies cannot be detected solely based on the detection information from the rotation angle detection device.

[0006] Therefore, the purpose of this application is to provide a rotation angle detection device that can determine the abnormality of the rotation angle detection device solely based on the detection information of the rotation angle detection device, even at low rotation speeds. Technical means for solving technical problems

[0007] The first rotation angle detection device of this application includes: There are N rotation detection sensors (N is an integer greater than or equal to 3). The output signal of each of these N sensors varies sinusoidally according to the rotation angle of the rotating component, and the sum of all sensor output signals is 0. An anomaly diagnosis unit is configured to diagnose the sensor in question, specifically the rotation detection sensor, which is used for anomaly diagnosis. The output values ​​of the N-1 rotation detection sensors other than the sensor for the diagnostic object are summed, and the value after reversing the sign of the sum is calculated as the estimated value of the sensor output value of the sensor for the diagnostic object. The difference between the sensor output value of the sensor for the diagnostic object and the estimated value of the sensor output value of the sensor for the diagnostic object is calculated as the estimated difference. If the estimated error is outside the range used for determining the estimated error, it is determined that an abnormality has occurred in the sensor of the diagnostic object.

[0008] The second rotation angle detection device of this application includes: There are N rotation detection sensors (N is an integer greater than or equal to 3). The output signal of each of these N sensors varies sinusoidally according to the rotation angle of the rotating component, and the sum of all sensor output signals is 0. An anomaly diagnosis unit is configured to diagnose the sensor in question, specifically the rotation detection sensor, which is used for anomaly diagnosis. The difference between the maximum and minimum values ​​of the sensor output of the diagnostic object sensor within one cycle of the sinusoidal sensor output signal, i.e., one cycle of the electrical angle, is the maximum-minimum difference. If the state where the magnitude of the maximum-minimum difference is below the judgment value used for the maximum-minimum difference persists for a fixed judgment period, it is determined that a fixed anomaly has occurred where the output signal of the diagnostic object sensor is fixed at a constant value.

[0009] The third rotation angle detection device of this application includes: There are N rotation detection sensors (N is an integer greater than or equal to 3). The output signal of each of these N sensors varies sinusoidally according to the rotation angle of the rotating component, and the sum of all sensor output signals is 0. An anomaly diagnosis unit is configured to diagnose the sensor in question, specifically the rotation detection sensor, which is used for anomaly diagnosis. Calculate the period difference between the sensor output value of the diagnostic object sensor before the difference period and the sensor output value of the diagnostic object sensor at this time. If the state of the period difference being within the fixed determination range including 0 continues for a fixed determination period, it is determined that a fixed anomaly has occurred in which the output signal of the diagnostic object sensor is fixed at a constant value.

[0010] The fourth rotation angle detection device of this application includes: There are N rotation detection sensors (N is an integer greater than or equal to 3). The output signal of each of these N sensors varies sinusoidally according to the rotation angle of the rotating component, and the sum of all sensor output signals is 0. An anomaly diagnosis unit is configured to diagnose the sensor in question, specifically the rotation detection sensor, which is used for anomaly diagnosis. Calculate the period difference between the sensor output value of the diagnostic object sensor before the difference period and the sensor output value of the diagnostic object sensor at this time. During the connection failure determination period, if the number of times the time difference is outside the connection failure determination range including 0 is greater than the number of connection failure determinations, it is determined that a connection failure has occurred in the signal line that transmits the output signal of the diagnostic object sensor. Invention Effects

[0011] According to the first rotation angle detection device of this application, under normal conditions, regardless of the rotation speed, when the total value of the sensor outputs of all sensors is 0, an estimated value of the sensor output of the diagnostic target sensor is calculated based on the sensor outputs of N-1 non-diagnostic target sensors. The occurrence of an abnormality in the diagnostic target sensor can be determined with high accuracy by determining whether the difference between the sensor output of the diagnostic target sensor and the estimated value (i.e., the estimated difference) is outside the range for determining the estimated error. Therefore, even at low rotation speeds, an abnormality in the rotation angle detection device can be determined solely by the detection information from the rotation angle detection device.

[0012] According to the second rotation angle detection device of this application, if a fixation abnormality occurs in the diagnostic object sensor, the difference between the maximum and minimum values ​​of the sensor output value of the diagnostic object sensor in the electrical angle 1 cycle (i.e., the maximum-minimum difference) is close to 0, thus enabling high-precision determination of the occurrence of the fixation abnormality. Therefore, even at low rotation speeds, the abnormality of the rotation angle detection device can be determined solely by the detection information from the rotation angle detection device.

[0013] According to the third rotation angle detection device of this application, if a fixation anomaly occurs in the sensor being diagnosed, the difference between the sensor output value of the sensor being diagnosed before the differential period and the sensor output value of the sensor being diagnosed now, i.e., the period difference, is close to 0. Therefore, the occurrence of a fixation anomaly can be determined with high accuracy. Furthermore, even under normal conditions, if the sensor output value is close to its maximum and minimum values, the period difference is close to 0. However, since the fixation determination period continues, a fixation anomaly will not be incorrectly determined. Thus, even at low rotation speeds, anomalies in the rotation angle detection device can be determined solely by the detection information from the rotation angle detection device.

[0014] According to the fourth rotation angle detection device of this application, if a signal line connection failure occurs, the output signal changes. Before and after the connection failure, the time difference between the sensor output value of the target sensor before the differential period and the sensor output value of the target sensor in the current period changes significantly relative to the normal vibration range centered at 0. Therefore, during the connection failure determination period, if the number of times the time difference is outside the determination range for connection failure exceeds the number of connection failure determinations, it can be determined that a signal line connection failure has occurred. Thus, even at low rotation speeds, an abnormality in the rotation angle detection device can be determined solely by the detection information from the rotation angle detection device. Attached Figure Description

[0015] Figure 1 This is a schematic structural diagram of the rotation angle detection device according to Embodiment 1. Figure 2 This is a simplified block diagram of the control device involved in Embodiment 1. Figure 3 This is a simplified hardware structure diagram of the control device involved in Implementation Method 1. Figure 4 This is a diagram illustrating the output signals of each rotation detection sensor involved in Embodiment 1. Figure 5 This is a flowchart illustrating the processing of the abnormality diagnosis unit involved in Implementation Method 1. Figure 6 This is a timing diagram illustrating the abnormality diagnosis actions when a fixed abnormality occurs, as described in Implementation Method 1. Figure 7 This is a timing diagram illustrating the abnormal diagnostic actions when a connection failure occurs, as described in Implementation Method 1. Figure 8 This is a flowchart illustrating the processing of the anomaly diagnosis unit involved in Embodiment 2. Figure 9 This is a timing diagram illustrating the detection of the electrical angle 1 cycle as described in Implementation Method 2. Figure 10 This is a timing diagram illustrating the abnormality diagnosis actions when a fixed abnormality occurs, as described in Implementation Method 2. Figure 11 This is a flowchart illustrating the processing of the anomaly diagnosis unit involved in Embodiment 3. Figure 12 This is a timing diagram illustrating the abnormality diagnosis actions when a fixed abnormality occurs, as described in Implementation Method 3. Figure 13 This is a flowchart illustrating the processing of the anomaly diagnosis unit involved in Embodiment 4. Figure 14This is a timing diagram illustrating the abnormal diagnostic actions when a connection failure occurs, as described in Implementation Method 4. Detailed Implementation

[0016] 1. Implementation Method 1 The rotation angle detection device 1 according to Embodiment 1 will be described with reference to the accompanying drawings. Figure 1 This is a simplified structural diagram of the rotation angle detection device 1 involved in this embodiment.

[0017] The rotation angle detection device 1 includes N rotation detection sensors 3 (N is an integer greater than or equal to 3) and a control device 30. In this embodiment, the rotation angle detection device 1 includes a rotor 2 and a magnetic field generating magnet 4.

[0018] 1-1. Rotor 2 The rotor 2 is fixed to a rotating member 9, such as the rotating shaft of an electric motor, and rotates integrally with the rotating member 9. In the rotor 2, M protrusions 2a (M is an integer greater than or equal to 1) are formed at equal intervals in the circumferential direction. The rotor 2 is composed of magnets. In this embodiment, M = 12. For every mechanical rotation of the rotor 2, the angle θ detected by the rotation angle detection device 1 rotates 12 times electrically.

[0019] The protrusion height of the protrusion 2a varies in a curved shape (in this example, a sine wave) in the circumferential direction, causing the output signal of the rotation detection sensor 3 to vary in a sine wave shape according to the rotation angle of the rotor 2. In this embodiment, the protrusion 2a is formed in the outer periphery of the cylindrical rotor 2, and the protrusion height toward the radially outward direction varies in a curved shape along the circumferential direction.

[0020] 1-2. Magnetic fields produce magnets 4 A magnetic field generating magnet 4 is disposed opposite to the protrusion 2a, generating a magnetic field between them. The magnetic field generating magnet 4 is disposed radially outside the protrusion 2a. The gap length between the protrusion 2a and the magnetic field generating magnet 4 varies according to the rotation angle of the rotor 2, and the magnetic flux density varies according to the gap length. The magnetic field generating magnet 4 is formed as an arc with a certain radius, the circumferential extension of which is longer than the circumferential extension of the N rotation detection sensors 3. The circumferential arrangement of the magnetic field generating magnet 4 is longer than the circumferential arrangement of the N rotation detection sensors 3, thus placing the magnetic field generating magnet 4 radially outside the N rotation detection sensors 3. Furthermore, the magnetic field generating magnet 4 can be divided circumferentially according to the circumferential arrangement of the N rotation detection sensors 3. The magnetic poles are different on the radially inner and radially outer sides of the magnetic field generating magnet 4, thereby generating radial magnetic flux. A magnetic core 5 is disposed on the radially outer side and circumferential sides of the magnetic field generating magnet 4, which has the effect of increasing the magnetic flux density of the gap. Alternatively, the magnetic core 5 can be omitted.

[0021] 1-3. Rotation detection sensor In the N rotation detection sensors 3, each output signal changes in a sinusoidal shape according to the rotation angle of the rotating body, and the sum of the output signals of all sensors becomes 0.

[0022] In this embodiment, N=3, and a first rotation detection sensor 3a, a second rotation detection sensor 3b, and a third rotation detection sensor 3c are provided. The sinusoidal output signals of the three rotation detection sensors 3 are arranged circumferentially, such that their phases differ by 360 / N degrees (120 degrees in this example) in electrical angle. 360 / N degrees (120 degrees in this example) in electrical angle corresponds to 360 / N / M degrees (10 degrees in this example) in mechanical angle. In this example, the three rotation detection sensors 3 are arranged circumferentially with a mechanical angle interval of 10 degrees. Alternatively, the three rotation detection sensors 3 can be arranged circumferentially such that their phases differ by 120 degrees in electrical angle, and the three rotation detection sensors 3 can be spaced apart by an electrical angle interval of 120 degrees + 360 degrees × 0 (where 0 is an integer greater than or equal to 1).

[0023] Each rotation detection sensor 3 is disposed radially inside the magnetic field generating magnet 4, and detects the magnetic flux density generated between the protrusion 2a and the magnetic field generating magnet 4. The magnetic flux density varies according to the distance between the protrusion 2a and the magnetic field generating magnet 4, and varies sinusoidally according to the rotation angle. Each rotation detection sensor 3 is a magnetic sensor for detecting magnetic flux density, and is a Hall element or the like. The output signal of each rotation detection sensor 3 is input to the control device 30.

[0024] 1-4. Control device 30 The rotation angle detection device 1 includes a control device 30. For example... Figure 2 As shown, the control device 30 includes a signal detection unit 31, an angle calculation unit 32, and an anomaly diagnosis unit 33.

[0025] Specifically, such as Figure 3 As shown, the control device 30 includes an arithmetic processing unit 90 such as a CPU (Central Processing Unit), a storage device 91, and an input / output device 92 for inputting and outputting external signals to the arithmetic processing unit 90.

[0026] As the arithmetic processing device 90, it can be equipped with ASIC (Application Specific Integrated Circuit), IC (Integrated Circuit), FPGA (Field Programmable Gate Array), various logic circuits, and various signal processing circuits. Furthermore, the arithmetic processing device 90 can also have multiple arithmetic processing devices of the same or different types to share the execution of each process. As the storage device 91, it can include RAM (Random Access Memory) configured to read and write data from the arithmetic processing device 90, ROM (Read Only Memory) configured to read data from the arithmetic processing device 90, etc.

[0027] The input / output device 92 includes a communication device, an A / D converter, input / output ports, and drive circuits. The input / output device 92 (A / D converter) is connected to each rotation detection sensor 3, performs A / D conversion on the output signals of each rotation detection sensor 3, and transmits the data to the processing unit 90. Furthermore, the input / output device 92 (communication device) is connected to an external control unit 95, and the rotation angle θ calculated by the processing unit 90 and the abnormal diagnostic information of the rotation detection sensors 3 are transmitted to the external control unit 95. The external control unit 95 is a control device for a rotating device such as a motor that rotates integrally with the rotating component 9.

[0028] Furthermore, the functions of each processing unit 31 to 33 in the control device 30 are realized by the arithmetic processing unit 90 executing software (programs) stored in a storage device 91 such as a ROM, and in cooperation with other hardware of the control device 30 such as the storage device 91 and the input / output device 92. Additionally, the setting data such as the determination range and determination value used by each processing unit 31 to 33 are stored as part of the software (program) in the storage device 91 such as a ROM. The functions of the control device 30 will be described in detail below.

[0029] <Signal Detection Unit 31> The signal detection unit 31 detects the sinusoidal sensor output components of the output signals of the three rotation detection sensors 3a, 3b, and 3c as sensor output values ​​SACa, SACb, and SACc. The signal detection unit 31 performs detection for each detection cycle.

[0030] like Figure 4As shown in the following formula, the output signals Soa, Sob, and Soc of each rotation detection sensor 3a, 3b, and 3c, in addition to the sinusoidal vibration component that varies with the rotation angle θ under the electrical angle of the rotating component, also have a DC offset component Off superimposed on them. [Mathematical Expression 1]

[0031] Here, Ko is the gain of the sensor output, which becomes the same value across sensors under ideal conditions without bias. The offset component Off also becomes the same value across sensors under ideal conditions without bias.

[0032] As shown in the following formula, the signal detection unit 31 calculates the sensor output values ​​SACa, SACb, and SACc of each sensor, which are sinusoidal sensor output components, by subtracting the offset component value Off from the output signals Soa, Sob, and Soc of each rotation detection sensor 3a, 3b, and 3c. [Mathematical Expression 2]

[0033] The offset component value Off can be a preset value. To address deviation factors, the value obtained through statistical processing, such as averaging the output signals of each rotation detection sensor (Soa, Sob, Soc) under normal conditions, can be set as the offset component value Off for each sensor. The offset component values ​​Off for each sensor can be different.

[0034] <Abnormal Diagnosis Department 33> Under normal conditions, as shown in the following formula, the total value of the sensor output values ​​SACa, SACb, and SACc of all sensors is 0. [Mathematical Expression 3]

[0035] As shown in equation (3), under normal conditions, the sensor output value of one sensor should be equal to the value after reversing the positive and negative values ​​of the sum of the sensor output values ​​of the other two sensors. [Mathematical Expression 4]

[0036] Therefore, the anomaly diagnosis unit 33 sets one rotational detection sensor, i.e., the diagnostic target sensor, for anomaly diagnosis. It sums the sensor output values ​​of N-1 other rotational detection sensors (hereinafter referred to as non-diagnostic target sensors), and calculates the value after reversing the sign of the summed value as the estimated value SACest of the diagnostic target sensor's sensor output value. Then, the anomaly diagnosis unit 33 calculates the estimated difference ΔSACest, which is the difference between the diagnostic target sensor's sensor output value SACest and the estimated value SACest. If the estimated difference ΔSACest is outside the judgment range used for the estimated difference, it is determined that the diagnostic target sensor has an anomaly.

[0037] The anomaly diagnosis unit 33 sequentially sets each of the rotation detection sensors 3a, 3b, and 3c as the diagnostic target sensor and determines the anomaly of each rotation detection sensor. For example, as shown in the following formula, when the diagnostic target sensor is set as the first rotation detection sensor 3a, the anomaly diagnosis unit 33 sums the sensor output values ​​SACb and SACc of the second rotation detection sensor 3b and the third rotation detection sensor 3c, and calculates the value obtained by reversing the sign of the summed value as the estimated value SACesta of the sensor output value of the first rotation detection sensor 3a. Then, the anomaly diagnosis unit 33 calculates the difference between the sensor output value SACa of the first rotation detection sensor 3a and the estimated value SACesta of the sensor output value of the first rotation detection sensor 3a, i.e., the estimated difference ΔSACesta. If the estimated difference ΔSACesta is outside the judgment range for the estimated error (from THestL to THestH), it is determined that an anomaly has occurred in the first rotation detection sensor 3a. [Mathematical Expression 5]

[0038] For example, the judgment range for the estimated error (from THestL to THestH) is set to a range of approximately a few percent of the vibration range of the sensor output value of the rotation detection sensor under normal conditions.

[0039] According to this structure, under normal conditions, taking advantage of the fact that the total value of the sensor output values ​​SACa, SACb, and SACc of all sensors is 0, the estimated value of the sensor output value of the diagnostic sensor is calculated based on the sensor output values ​​of N-1 non-diagnostic sensors. By determining whether the difference between the sensor output value of the diagnostic sensor and the estimated value of the diagnostic sensor, i.e., the estimated difference ΔSACest, is outside the judgment range used for the estimated error, the occurrence of abnormality of the diagnostic sensor can be determined with high precision.

[0040] The anomaly diagnosis unit 33 transmits the anomaly determination information to the angle calculation unit 32 and external control devices 95, etc.

[0041] <Flowchart> The processing of the abnormality diagnosis unit 33 in this embodiment is as follows: Figure 5 It is structured as shown in the flowchart. Execution is performed for each computation cycle. Figure 5 The processing of flowcharts.

[0042] In step S01, the anomaly diagnosis unit 33 sets the target sensor for diagnosis. For example, in each operation cycle, the anomaly diagnosis unit 33 sequentially sets each rotation detection sensor 3a, 3b, and 3c as the target sensor for diagnosis.

[0043] In step S02, the anomaly diagnosis unit 33 determines whether the sensor output values ​​of all N rotation detection sensors are within the normal range. For each of the N rotation detection sensors, the anomaly diagnosis unit 33 determines whether the sensor output value is within the normal range; for all N rotation detection sensors, it determines whether they are within the normal range. Then, if all are determined to be within the normal range, the process proceeds to step S03; otherwise, it proceeds to step S11. In step S11, the anomaly diagnosis unit 33 determines that an anomaly has occurred in rotation detection sensor 3 (the rotation detection sensor not within the normal range), and the process ends.

[0044] The normal range is set to include the range of changes in the sensor output value under normal conditions. For example, the normal range is set to be approximately 10% wider than the normal range of change.

[0045] Based on this structure, it is possible to use a rotation detection sensor whose sensor output value is outside the normal range to prevent abnormal judgment of the diagnostic object sensor, thereby improving the accuracy of abnormal judgment. Furthermore, it can determine that an abnormality has occurred in a rotation detection sensor whose sensor output value is outside the normal range.

[0046] In step S03, as described above, the anomaly diagnosis unit 33 sums the sensor output values ​​of N-1 non-diagnostic sensors and calculates the value after reversing the sign of the sum as the estimated value SACest of the sensor output value of the diagnostic sensor. Then, the anomaly diagnosis unit 33 calculates the estimated difference ΔSACest, which is the difference between the sensor output value SAC of the diagnostic sensor and the estimated value SACest of the sensor output value of the diagnostic sensor.

[0047] Then, in step S04, as described above, the anomaly diagnosis unit 33 determines whether the estimated difference ΔSACest is outside the determination range used for estimated difference. If it is not outside the determination range used for estimated difference, it proceeds to step S05; if it is outside the determination range used for estimated difference, it proceeds to step S06. In step S05, the anomaly diagnosis unit 33 determines that the sensor being diagnosed is normal and ends the process.

[0048] In step S06, as shown in the following formula, the abnormality diagnosis unit 33 determines whether the difference between the sensor output value SACold of the diagnostic target sensor before the differential period ΔTdt and the sensor output value SAC of the current diagnostic target sensor, i.e., the period difference ΔSACt, is within the fixed determination range (THfxL to THfxH) including 0. If it is within the fixed determination range, the process proceeds to step S07; otherwise, it proceeds to step S08. [Mathematical Expression 6] ΔSACt=SAC-SACold…(6) THfxL≤ΔSACt≤THfxH

[0049] In this embodiment, the differential period ΔTdt is set as the detection period, using the sensor output value SACold of the diagnostic object sensor detected at the last detection timing. The differential period ΔTdt can also be set as an integer multiple of 2 or more of the detection period.

[0050] Then, in step S07, since the estimated difference ΔSACest is outside the judgment range for estimated difference and the period difference ΔSACt is within the judgment range for fixed determination, the anomaly diagnosis unit 33 determines that a fixed anomaly has occurred where the output signal of the diagnostic target sensor is fixed at a constant value.

[0051] When a fixed anomaly occurs, the output signal becomes a constant value, and the period difference ΔSACt becomes close to 0. Therefore, when the period difference ΔSACt is within the fixed determination range including 0, a fixed anomaly can be determined to have occurred even among the anomaly types.

[0052] On the other hand, in step S08, the anomaly diagnosis unit 33 determines whether the estimated difference ΔSACest is outside the connection determination range (THcnL to THcnH), which is wider than the determination range used for the estimated difference. If it is outside the connection determination range, the process proceeds to step S09; if it is not outside the connection determination range, the process proceeds to step S10. For example, the connection determination range (THcnL to THcnH) is set to approximately 10% of the sensor output range of the rotation detection sensor under normal conditions.

[0053] In step S09, since the estimated difference ΔSACest is outside the judgment range for estimated difference and outside the judgment range for connection judgment, the abnormality diagnosis unit 33 determines that a connection failure has occurred in the signal line that transmits the output signal of the sensor of the diagnostic object.

[0054] When a signal line connection failure occurs, vibration or other factors can cause the output signal to fluctuate intermittently or continuously, resulting in a large or intermittent fluctuation in the estimated error ΔSACest. Therefore, even if the estimated error ΔSACest falls outside the range of connection determination, which is wider than the range used for estimation error, a signal line connection failure can still be determined to have occurred, even if it falls within the category of anomalies.

[0055] On the other hand, in step S10, since the estimated difference ΔSACest is outside the judgment range used for the estimated difference, and some kind of abnormality other than poor connection and fixing abnormality has occurred, the abnormality diagnosis unit 33 simply determines that an abnormality has occurred in the sensor of the diagnostic object and ends the process.

[0056] <Examples of action determination> Figure 6 The determination action is shown when a fixation anomaly occurs in the first rotation detection sensor 3a. Before time t01, the first rotation detection sensor 3a is normal, and after time t01, a fixation anomaly occurs, and the sensor output value SACa of the first rotation detection sensor 3a is fixed to a constant value.

[0057] Before time t01, the sensor output value SACa of the first rotation detection sensor 3a is consistent with the estimated value SACesta of the first rotation detection sensor 3a, and the estimated error ΔSACesta is close to 0, and within the judgment range (THestL to THEstH) used for the estimated error. Therefore, the first rotation detection sensor 3a is judged to be normal. However, after time t01, since it is fixed, the estimated error ΔSACesta changes from near 0 and is outside the judgment range (THestL to THEstH) used for the estimated error, and it is judged that an anomaly has occurred in the first rotation detection sensor 3a.

[0058] Before time t01, the period difference ΔSACta, which is the difference between the previous sensor output value SAColda and the current sensor output value SACa, fluctuates periodically. After the fixed time t01, the period difference ΔSACta becomes 0 and falls within the fixed judgment range (THfxL to THfxH). Therefore, even among the types of anomalies, it is possible to accurately determine that a fixed anomaly has occurred.

[0059] Figure 7The determination action is shown when a connection failure occurs in the first rotation detection sensor 3a. Before time t02, the first rotation detection sensor 3a is normal; after time t02, a connection failure occurs, and the sensor output value SACa of the first rotation detection sensor 3a changes intermittently.

[0060] Before time t02, the sensor output value SACa of the first rotation detection sensor 3a was consistent with the estimated value SACesta of the first rotation detection sensor 3a, and the estimated difference ΔSACesta was near 0, within the judgment range (THestL to THEstH) used for the estimated error. Therefore, the first rotation detection sensor 3a was judged to be normal. However, after time t02, due to poor connection, the estimated difference ΔSACesta fluctuated intermittently from near 0, and the estimated difference ΔSACesta was outside the judgment range (THestL to THEstH) used for the estimated error, indicating that an anomaly had occurred in the first rotation detection sensor 3a.

[0061] Due to the occurrence of a poor connection, the estimated error ΔSACesta changes significantly from 0, falling outside the range used for connection determination (THcnL to THcnH), which is wider than the range used for estimated error (THestL to THestH). Therefore, even among the types of anomalies, a poor connection can be determined with high accuracy.

[0062] <Angle Calculation Section 32> The angle calculation unit 32 calculates the rotation angle θ of the rotating member 9 based on the sensor output values ​​of the three rotation detection sensors 3. In this embodiment, if no abnormality is determined to have occurred in the rotation detection sensors 3, the angle calculation unit 32 calculates the rotation angle θ at the electrical angle based on the sensor output values ​​SACa, SACb, and SACc of the three rotation detection sensors 3 using the following formula. [Mathematical Expression 7]

[0063] On the other hand, if an abnormality is determined to have occurred in the rotation detection sensor 3, the angle calculation unit 32 calculates the rotation angle θ based on the sensor output value SAC of the rotation detection sensor that is not determined to have occurred. In this embodiment, as shown in equations (4) and (5), the angle calculation unit 32 sums the sensor output values ​​of N-1 normal rotation detection sensors other than the abnormal rotation detection sensor, and calculates the value after reversing the sign of the sum as the estimated value SACest of the sensor output value of the abnormal rotation detection sensor. Then, in equation (7), the angle calculation unit 32 uses the estimated value SACest of the sensor output value of the abnormal rotation detection sensor instead of the sensor output value SAC of the abnormal rotation detection sensor, and calculates the rotation angle θ under the electrical angle based on the sensor output values ​​SAC of N-1 normal rotation detection sensors and the estimated value SACest of the sensor output value of the abnormal rotation detection sensor.

[0064] 2. Implementation Method 2 The rotation angle detection device 1 according to Embodiment 2 will be described with reference to the accompanying drawings. Descriptions of structural parts identical to those in Embodiment 1 are omitted. The basic structure of the rotation angle detection device 1 according to this embodiment is the same as that in Embodiment 1. However, the difference in this embodiment compared to Embodiment 1 lies in the processing of the abnormality diagnosis unit 33.

[0065] In this embodiment, the anomaly diagnosis unit 33 sets a rotating detection sensor, i.e., the diagnostic target sensor, to perform anomaly diagnosis. As shown in the following formula, the difference between the maximum value SACmax and the minimum value SACmin of the sensor output value SAC of the diagnostic target sensor in one period of the sinusoidal sensor output value, i.e., the maximum-minimum difference ΔSACmnmx, is calculated. If the magnitude of the maximum-minimum difference ΔSACmnmx is below the determination value THmnmx used for the maximum-minimum difference and remains fixed for a determination period ΔTfx, it is determined that a fixed anomaly has occurred in which the output signal of the diagnostic target sensor is fixed at a constant value. [Mathematical Expression 8] ΔSACmnmx=SACmax-SACmin…(8) ΔSACmnmx≤THmnmx

[0066] If a fixed anomaly occurs, the maximum and minimum difference between the electrical angle 1-period ΔTcyl and ΔSACmnmx will be close to 0. Therefore, based on the above structure, the occurrence of a fixed anomaly can be determined with high accuracy.

[0067] <Flowchart> The processing of the abnormality diagnosis unit 33 in this embodiment is as follows: Figure 8It is structured as shown in the flowchart. Execution is performed for each computation cycle. Figure 8 The processing of flowcharts.

[0068] In step S21, the anomaly diagnosis unit 33 sets the diagnostic target sensor. For example, in each operation cycle, the anomaly diagnosis unit 33 sequentially sets each rotation detection sensor 3a, 3b, and 3c as the diagnostic target sensor.

[0069] In step S22, the anomaly diagnosis unit 33 calculates the electrical angle 1-cycle ΔTcyl based on the moment when the sensor output value SAC of the non-diagnostic sensor other than the diagnostic sensor crosses the center value (0 in this example).

[0070] like Figure 9 As shown, the anomaly diagnosis unit 33 uses a timer to measure the moment when the sensor output value SAC of the non-diagnostic object sensor crosses 0, and calculates the time difference between the previous crossing moment and the current crossing moment as the electrical angle 1-cycle ΔTcyl. At this time, the crossing moments detected in past calculation cycles are used. The calculated electrical angle 1-cycle ΔTcyl can be averaged.

[0071] Based on this structure, the electrical angle period ΔTcyl can be calculated with high precision based on the sensor output value SAC of the non-diagnostic object sensor.

[0072] In step S23, the anomaly diagnosis unit 33 sets the determination value THmnmx for the maximum and minimum difference and the fixed determination period ΔTfx based on the electrical angle 1-cycle ΔTcyl. As the electrical angle 1-cycle ΔTcyl increases, the anomaly diagnosis unit 33 decreases the determination value THmnmx for the maximum and minimum difference and increases the fixed determination period ΔTfx.

[0073] Since the period required to calculate the maximum and minimum difference ΔSACmnmx varies proportionally to the electrical angle 1-cycle ΔTcyl, the judgment value THmnmx for the maximum and minimum difference and the fixed judgment period ΔTfx can be appropriately set according to the electrical angle 1-cycle ΔTcyl.

[0074] In step S24, the anomaly diagnosis unit 33 calculates the maximum value SACmax and minimum value SACmin of the sensor output value SAC of the target sensor during the electrical angle 1 period ΔTcyl. For example, the anomaly diagnosis unit 33 determines the maximum value SACmax and minimum value SACmin from the sensor output values ​​SAC of the target sensor detected during the past electrical angle 1 period ΔTcyl. Then, the anomaly diagnosis unit 33 subtracts the minimum value SACmin from the maximum value SACmax to calculate the maximum-minimum difference ΔSACmnmx.

[0075] In step S25, the anomaly diagnosis unit 33 determines whether the maximum-minimum difference ΔSACmnmx is above the determination value THmnmx used for the maximum-minimum difference. If it is above the determination value THmnmx used for the maximum-minimum difference, it proceeds to step S27. If it is not above the determination value THmnmx used for the maximum-minimum difference, it proceeds to step S26.

[0076] In step S26, the anomaly diagnosis unit 33 decrements the anomaly determination counter Cnt by 1. Furthermore, the anomaly determination counter Cnt is limited to a lower limit of 0, with an initial value of 0. On the other hand, in step S27, the anomaly diagnosis unit 33 increments the anomaly determination counter Cnt by 1.

[0077] In step S28, the anomaly diagnosis unit 33 determines whether the anomaly determination counter Cnt is above the fixed determination period ΔTfx. If it is above the fixed determination period ΔTfx, it proceeds to step S30; otherwise, it proceeds to step S29.

[0078] In step S29, the anomaly diagnosis unit 33 determines that the sensor being diagnosed is normal (or that no fixed anomaly has occurred in the sensor being diagnosed). On the other hand, in step S30, the anomaly diagnosis unit 33 determines that a fixed anomaly has occurred in the sensor being diagnosed.

[0079] <Examples of action determination> Figure 10 The determination action is shown when a fixation anomaly occurs in the first rotation detection sensor 3a. Before time t11, the first rotation detection sensor 3a is normal, and after time t11, a fixation anomaly occurs, and the sensor output value SACa of the first rotation detection sensor 3a is fixed to a constant value.

[0080] exist Figure 10 In the example, the maximum and minimum difference ΔSACmnmx is calculated for each electrical angle 1-cycle ΔTcyl. Since there is no fixation before time t11, the maximum and minimum difference ΔSACmnmx exceeds the determination value THmnmx used for the maximum and minimum difference. However, after time t11, due to a fixation anomaly, the maximum and minimum difference ΔSACmnmx decreases to 0, which is less than the determination value THmnmx used for the maximum and minimum difference. Although not shown in the figure, when the state of the maximum and minimum difference ΔSACmnmx being less than the determination value THmnmx used for the maximum and minimum difference persists for a fixation period ΔTfx, it is determined that a fixation anomaly has occurred in the first rotation detection sensor 3a.

[0081] 3. Implementation Method 3 The rotation angle detection device 1 according to Embodiment 3 will be described with reference to the accompanying drawings. Descriptions of structural parts identical to those in Embodiment 1 are omitted. The basic structure of the rotation angle detection device 1 according to this embodiment is the same as that in Embodiment 1. However, the difference between this embodiment and Embodiment 1 lies in the processing of the abnormality diagnosis unit 33.

[0082] In this embodiment, the anomaly diagnosis unit 33 sets a rotating detection sensor, namely the diagnostic target sensor, to perform anomaly diagnosis. As shown in the following formula, the difference between the sensor output value SACold of the diagnostic target sensor before the differential period ΔTdt and the sensor output value SAC of the current diagnostic target sensor is calculated, namely the period difference ΔSACt. If the period difference ΔSACt is within the fixed determination range (THfxL to THfxH) including 0 for a fixed determination period ΔTfx, it is determined that a fixed anomaly has occurred in which the output signal of the diagnostic target sensor is fixed at a constant value. [Mathematical Expression 9] ΔSACt=SAC-SACold…(9) THfxL≤ΔSACt≤THfxH

[0083] In this embodiment, the differential period ΔTdt is set as the detection period, using the sensor output value SACold of the diagnostic object sensor detected in the last detection timing. The differential period ΔTdt can be set to an integer multiple of 2 or more of the detection period.

[0084] If a fixed anomaly occurs, the period difference ΔSACt approaches 0. Therefore, based on the above structure, the occurrence of a fixed anomaly can be determined with high accuracy. Furthermore, even under normal conditions, if the sensor output value SAC approaches its maximum and minimum values, the period difference ΔSACt approaches 0. However, since the fixed determination period ΔTfx continues, it will not be incorrectly determined that a fixed anomaly has occurred.

[0085] <Flowchart> The processing of the abnormality diagnosis unit 33 in this embodiment is as follows: Figure 11 It is structured as shown in the flowchart. Execution is performed for each computation cycle. Figure 11 The processing of flowcharts.

[0086] In step S41, the anomaly diagnosis unit 33 sets the diagnostic target sensor. For example, in each operation cycle, the anomaly diagnosis unit 33 sequentially sets each rotation detection sensor 3a, 3b, and 3c as the diagnostic target sensor.

[0087] In step S42, the anomaly diagnosis unit 33 calculates the electrical angle period ΔTcyl based on the moment when the sensor output value SAC of the non-diagnostic sensor (other than the diagnostic sensor) crosses the center value (0 in this example). Since this is the same process as step S22 in Embodiment 2, the explanation is omitted.

[0088] In step S43, the anomaly diagnosis unit 33 sets a fixed judgment range (THfxL to THfxH) and a fixed judgment period ΔTfx based on the electrical angle 1-cycle ΔTcyl. As the electrical angle 1-cycle ΔTcyl increases, the anomaly diagnosis unit 33 reduces the fixed judgment range (THfxL to THfxH) and increases the fixed judgment period ΔTfx.

[0089] If the electrical angle period 1 ΔTcyl increases, the normal period difference ΔSACt decreases, and the period during which the normal period difference ΔSACt approaches 0 near its maximum or minimum value becomes longer. Therefore, as the electrical angle period 1 ΔTcyl increases, the judgment range used for fixed judgment (THfxL to THfxH) is reduced, and the fixed judgment period ΔTfx is increased, thereby maintaining judgment accuracy.

[0090] In step S44, the anomaly diagnosis unit 33 calculates the period difference ΔSACt, which is the difference between the sensor output value SACold of the diagnostic target sensor before the difference period ΔTdt and the sensor output value SAC of the current diagnostic target sensor.

[0091] In step S45, the abnormality diagnosis unit 33 determines whether the time difference ΔSACt is within the fixed determination range (THfxL to THfxH). If it is within the fixed determination range, the process proceeds to step S47; otherwise, the process proceeds to step S46.

[0092] In step S46, the anomaly diagnosis unit 33 decrements the anomaly determination counter Cnt by 1. Furthermore, the anomaly determination counter Cnt is limited to a lower limit of 0, with an initial value of 0. Conversely, in step S47, the anomaly diagnosis unit 33 increments the anomaly determination counter Cnt by 1.

[0093] In step S48, the anomaly diagnosis unit 33 determines whether the anomaly determination counter Cnt is above the fixed determination period ΔTfx. If it is above the fixed determination period ΔTfx, it proceeds to step S50; otherwise, it proceeds to step S49.

[0094] In step S49, the anomaly diagnosis unit 33 determines that the sensor being diagnosed is normal (or that no fixed anomaly has occurred in the sensor being diagnosed). On the other hand, in step S50, the anomaly diagnosis unit 33 determines that a fixed anomaly has occurred in the sensor being diagnosed.

[0095] <Examples of action determination> Figure 12 The determination action is shown when a fixation anomaly occurs in the first rotation detection sensor 3a. Before time t21, the first rotation detection sensor 3a is normal, and after time t21, a fixation anomaly occurs, and the sensor output value SACa of the first rotation detection sensor 3a is fixed to a constant value.

[0096] Since there is no fixed value before time t21, the period difference ΔSACt varies with an electrical angle period of 1. Even under normal conditions, if the sensor output value SACa is close to the maximum and minimum values, the period difference ΔSACta is close to 0, which is within the determination range (THfxL to THfxH) used for fixing. However, since the determination period ΔTfx continues, there is no false determination that a fixing anomaly has occurred.

[0097] However, after time t21, due to a fixation anomaly, the period difference ΔSACta becomes 0 and falls within the fixation determination range (THfxL to THfxH). Moreover, although not shown in the figure, when the period difference ΔSACta remains within the fixation determination range for a fixed determination period ΔTfx, it is determined that a fixation anomaly has occurred in the first rotation detection sensor 3a.

[0098] 4. Implementation Method 4 The rotation angle detection device 1 according to Embodiment 4 will be described with reference to the accompanying drawings. Descriptions of structural parts identical to those in Embodiment 1 are omitted. The basic structure of the rotation angle detection device 1 according to this embodiment is the same as that in Embodiment 1. However, the difference between this embodiment and Embodiment 1 lies in the processing of the abnormality diagnosis unit 33.

[0099] In this embodiment, the anomaly diagnosis unit 33 sets a rotating detection sensor, namely the diagnostic target sensor, to perform anomaly diagnosis. As shown in the following formula, the difference between the sensor output value SACold of the diagnostic target sensor before the differential period ΔTdt and the sensor output value SAC of the current diagnostic target sensor is calculated, namely the period difference ΔSACt. During the connection failure determination period ΔTcn, when the number of times the period difference ΔSACt becomes outside the connection failure determination range (THcnL to THcnH) containing 0 reaches more than the connection failure determination number Ncn, it is determined that a connection failure has occurred in the signal line that transmits the output signal of the diagnostic target sensor. [Mathematical Expression 10] ΔSACt=SAC-SACold…(10) ΔSACt≤THcnL,THcnH<ΔSACt

[0100] In this embodiment, the differential period ΔTdt is set as the detection period, using the sensor output value SACold of the diagnostic object sensor detected in the last detection timing. The differential period ΔTdt can be set to an integer multiple of 2 or more of the detection period.

[0101] If a signal line connection is faulty, the output signal will fluctuate. Before and after the faulty connection, the time difference ΔSACt changes significantly relative to the normal sensor output range centered at 0. Therefore, during the faulty connection determination period ΔTcn, if the number of times the time difference ΔSACt falls outside the faulty connection determination range exceeds the number of faulty connection determinations Ncn, a faulty signal line connection can be determined.

[0102] <Flowchart> The processing of the abnormality diagnosis unit 33 in this embodiment is as follows: Figure 13 It is structured as shown in the flowchart. Execution is performed for each computation cycle. Figure 13 The processing of flowcharts.

[0103] In step S61, the anomaly diagnosis unit 33 sets the target sensor for diagnosis. For example, in each operation cycle, the anomaly diagnosis unit 33 sequentially sets each rotation detection sensor 3a, 3b, and 3c as the target sensor for diagnosis.

[0104] In step S62, the anomaly diagnosis unit 33 calculates the electrical angle period ΔTcyl based on the moment when the sensor output value SAC of the non-diagnostic sensor (other than the diagnostic target sensor) crosses the center value (0 in this example). Since this is the same process as step S22 in Embodiment 2, its description is omitted.

[0105] In step S63, the anomaly diagnosis unit 33 sets the judgment range (THcnL to THcnH) and the judgment period ΔTcn for connection failure based on the electrical angle 1-cycle ΔTcyl. As the electrical angle 1-cycle ΔTcyl increases, the anomaly diagnosis unit 33 reduces the judgment range (THcnL to THcnH) for connection failure and increases the judgment period ΔTcn for connection failure.

[0106] If the electrical angle period 1 ΔTcyl increases, the sensor output range of the normal period difference ΔSACt narrows, and the variation period of the normal period difference ΔSACt increases. Therefore, as the electrical angle period 1 ΔTcyl increases, the judgment range for connection failure (THcnL to THcnH) decreases due to the narrowing of the sensor output range of the normal period difference ΔSACt, and the connection failure judgment period ΔTcn increases due to the increase in the variation period of the normal period difference ΔSACt, thus maintaining judgment accuracy.

[0107] In step S64, the anomaly diagnosis unit 33 calculates the period difference ΔSACt, which is the difference between the sensor output value SACold of the diagnostic target sensor before the difference period ΔTdt and the sensor output value SAC of the current diagnostic target sensor.

[0108] In step S65, the anomaly diagnosis unit 33 determines whether the time difference ΔSACt is outside the determination range (THcnL to THcnH) for connection failure. If it is outside the determination range for connection failure, the process proceeds to step S67. If it is not outside the determination range for connection failure, the process proceeds to step S66.

[0109] In step S66, the anomaly diagnosis unit 33 sets the determination result Rlt of the current operation cycle to 0, indicating that it is not outside the determination range. On the other hand, in step S67, the anomaly diagnosis unit 33 sets the determination result Rlt of the current operation cycle to 1, indicating that it is outside the determination range.

[0110] In step S68, the anomaly diagnosis unit 33 counts the number of times the judgment result Rlt is set to 1 during the past connection failure judgment period ΔTcn, and determines whether the number of times it is set to 1 is greater than or equal to the number of connection failure judgments Ncn. If the number of connection failure judgments Ncn is greater than or equal to the number of connection failure judgments Ncn, the process proceeds to step S70. If the number of connection failure judgments Ncn is not greater than or equal to the number of connection failure judgments Ncn, the process proceeds to step S69.

[0111] In step S69, the anomaly diagnosis unit 33 determines that the sensor being diagnosed is normal (or that there is no connection failure in the sensor being diagnosed). On the other hand, in step S70, the anomaly diagnosis unit 33 determines that a connection failure has occurred in the sensor being diagnosed.

[0112] <Examples of action determination> Figure 14 The determination action is shown when a fixed abnormality occurs in the first rotation detection sensor 3a. Before time t31, the first rotation detection sensor 3a is normal. After time t31, a connection failure occurs, and the sensor output value SACa of the first rotation detection sensor 3a changes intermittently.

[0113] Since no connection failure occurred before time t31, the period difference ΔSACta varied within the range for determining connection failure (THcnL to THcnH) with an electrical angle period of 1. However, after time t31, connection failures occurred intermittently. Before and after the occurrence of connection failures, the period difference ΔSACta varied significantly relative to the normal sensor output range centered at 0, falling outside the range for determining connection failure (THcnL to THcnH).

[0114] During the connection failure determination period ΔTcn, when the number of times the period difference ΔSACta is outside the connection failure determination range (THcnL to THcnH) exceeds the connection failure determination number Ncn, a connection failure is determined to have occurred in the first rotation detection sensor 3a.

[0115] <Example> In the above embodiments, a magnetic sensor is used as the rotation detection sensor 3. However, any type of sensor can be used as long as there are N rotation detection sensors whose output values ​​change sinusoidally according to the rotation angle of the rotating member and whose total output signal of all sensors is 0. Furthermore, N does not have to be 3; N can be 4 or more.

[0116] In the above embodiments, the electrical angle period ΔTcyl is calculated based on the moment when the sensor output value SAC of the non-diagnostic sensor crosses the center value (0). However, the electrical angle period ΔTcyl can also be set based on the zero-crossing moment of the current detection value of the motor winding detected by the motor control device, the zero-crossing moment of the induced voltage of the motor winding, or speed information, etc.

[0117] The control device 30 can be integrated with the control device of a rotating device such as an electric motor that rotates integrally with the rotating component 9.

[0118] While this application describes various exemplary embodiments and examples, the various features, methods, and functions described in one or more embodiments are not limited to the application of a particular embodiment and can be applied to the embodiments individually or in various combinations. Therefore, it can be considered that numerous modifications not illustrated are also included within the scope of the technology disclosed in this application. For example, this includes cases where at least one constituent element is modified, added to, or omitted, and cases where at least one constituent element is extracted and combined with constituent elements of other embodiments. Label Explanation

[0119] 1 Rotation angle detection device, 3 Rotation detection sensor, 9 Rotation component, 31 Signal detection unit, 32 Angle calculation unit, 33 Anomaly diagnosis unit, Ncn Number of times to determine poor connection, SAC sensor output value, SACest estimated value, SACmax maximum value, SACmin minimum value, SACold Sensor output value before differential period, THmnmx Judgment value for maximum and minimum difference, ΔSACest estimated difference, ΔSACmnmx maximum and minimum difference, ΔSACt period difference, ΔTcn poor connection judgment period, ΔTcyl electrical angle 1 cycle, ΔTdt differential period, ΔTfx fixed judgment period, θ rotation angle.

Claims

1. A rotation angle detection device, characterized in that, include: There are N rotation detection sensors, each of which outputs a sinusoidal signal that varies according to the rotation angle of the rotating component, and the sum of all sensor outputs is 0, where N is an integer greater than or equal to 3; An anomaly diagnosis unit is configured to diagnose the sensor in question, specifically the rotation detection sensor, which is used for anomaly diagnosis. The output values ​​of the N-1 rotation detection sensors other than the sensor for the diagnostic object are summed, and the value after reversing the sign of the sum is calculated as the estimated value of the sensor output value of the sensor for the diagnostic object. The difference between the sensor output value of the sensor for the diagnostic object and the estimated value of the sensor output value of the sensor for the diagnostic object is calculated as the estimated difference. If the estimated error is outside the range used for determining the estimated error, it is determined that an abnormality has occurred in the sensor of the diagnostic object.

2. The rotation angle detection device as described in claim 1, characterized in that, If the estimated difference is outside the judgment range used for the estimated difference, and the difference between the sensor output value of the diagnostic object sensor before the difference period and the sensor output value of the diagnostic object sensor at this time, i.e., the period difference, is within the judgment range used for fixed judgment including 0, then the abnormality diagnosis unit determines that a fixed abnormality has occurred in which the output signal of the diagnostic object sensor is fixed at a constant value.

3. The rotation angle detection device as described in claim 1 or 2, characterized in that, The anomaly diagnosis unit determines that a faulty connection has occurred in the signal line used to transmit the output signal of the sensor being diagnosed, when the estimated difference is outside the judgment range used for the estimated difference and the estimated difference is outside the judgment range used for connection determination which is larger than the judgment range used for the estimated difference.

4. The rotation angle detection device as described in claim 1 or 2, characterized in that, When all the sensor output values ​​of the N rotation detection sensors are within the normal range, the anomaly diagnosis unit uses the estimated value of the sensor output value of the target sensor to determine the anomaly of the target sensor. When any one of the sensor output values ​​of the N rotation detection sensors is outside the normal range, the estimated value of the sensor output value of the target sensor is not used to determine the anomaly of the target sensor.

5. The rotation angle detection device as described in claim 1 or 2, characterized in that, The system includes an angle calculation unit that calculates the rotation angle of the rotating component based on the sensor output values ​​of the N rotation detection sensors. If an abnormality is determined to have occurred in the sensor of the diagnostic object, the angle calculation unit calculates the rotation angle based on the sensor output value of the rotation detection sensor that is not determined to have occurred.

6. A rotation angle detection device, characterized in that, include: There are N rotation detection sensors, each of which outputs a sinusoidal signal that varies according to the rotation angle of the rotating component, and the sum of all sensor outputs is 0, where N is an integer greater than or equal to 3; An anomaly diagnosis unit is configured to diagnose the sensor in question, specifically the rotation detection sensor, which is used for anomaly diagnosis. The difference between the maximum and minimum values ​​of the sensor output value of the diagnostic object sensor is calculated within one cycle of the sinusoidal sensor output value, i.e., one cycle of the electrical angle. If the state where the magnitude of the maximum-minimum difference is below the judgment value used for the maximum-minimum difference persists for a fixed judgment period, it is determined that a fixed anomaly has occurred where the output signal of the diagnostic object sensor remains at a constant value. The abnormality diagnosis unit calculates the electrical angle period 1 based on the moment when the sensor output value of the rotation detection sensor other than the diagnostic object sensor crosses the center value. As the electrical angle period 1 becomes longer, the determination value of the maximum and minimum difference is reduced, and the fixed determination period is increased.

7. The rotation angle detection device as described in claim 6, characterized in that, The system includes an angle calculation unit that calculates the rotation angle of the rotating component based on the sensor output values ​​of the N rotation detection sensors. If an abnormality is determined to have occurred in the sensor of the diagnostic object, the angle calculation unit calculates the rotation angle based on the sensor output value of the rotation detection sensor that is not determined to have occurred.

8. A rotation angle detection device, characterized in that, include: There are N rotation detection sensors, each of which outputs a sinusoidal signal that varies according to the rotation angle of the rotating component, and the sum of all sensor outputs is 0, where N is an integer greater than or equal to 3; An anomaly diagnosis unit is configured to diagnose the sensor in question, specifically the rotation detection sensor, which is used for anomaly diagnosis. Calculate the period difference between the sensor output value of the diagnostic object sensor before the difference period and the sensor output value of the diagnostic object sensor at this time. If the state where the period difference is within a fixed determination range including 0 persists for a fixed determination period, it is determined that a fixed anomaly has occurred, where the output signal of the diagnostic sensor remains a constant value. The abnormality diagnosis unit calculates one cycle of the sinusoidal sensor output value, namely the electrical angle 1 cycle, based on the moment when the sensor output value of the rotation detection sensor other than the diagnostic object sensor crosses the center value. As the electrical angle 1 cycle becomes longer, the judgment range for the fixed judgment is reduced, and the fixed judgment period is increased.

9. The rotation angle detection device as described in claim 8, characterized in that, The system includes an angle calculation unit that calculates the rotation angle of the rotating component based on the sensor output values ​​of the N rotation detection sensors. If an abnormality is determined to have occurred in the sensor of the diagnostic object, the angle calculation unit calculates the rotation angle based on the sensor output value of the rotation detection sensor that is not determined to have occurred.

10. A rotation angle detection device, characterized in that, include: There are N rotation detection sensors, each of which outputs a sinusoidal signal that varies according to the rotation angle of the rotating component, and the sum of all sensor outputs is 0, where N is an integer greater than or equal to 3; An anomaly diagnosis unit is configured to diagnose the sensor in question, specifically the rotation detection sensor, which is used for anomaly diagnosis. Calculate the period difference between the sensor output value of the diagnostic object sensor before the difference period and the sensor output value of the diagnostic object sensor at this time. During the connection failure determination period, if the number of times the period difference is outside the connection failure determination range (including 0) exceeds the number of connection failure determinations, it is determined that a connection failure has occurred on the signal line transmitting the output signal of the sensor being diagnosed. The abnormality diagnosis unit calculates one cycle of the sinusoidal sensor output value, namely the electrical angle 1 cycle, based on the moment when the sensor output value of the rotation detection sensor other than the sensor of the diagnostic object crosses the center value. As the electrical angle 1 cycle becomes longer, the judgment range of the connection failure is narrowed, and the judgment period of the connection failure is increased.

11. The rotation angle detection device as described in claim 10, characterized in that, The system includes an angle calculation unit that calculates the rotation angle of the rotating component based on the sensor output values ​​of the N rotation detection sensors. If an abnormality is determined to have occurred in the sensor of the diagnostic object, the angle calculation unit calculates the rotation angle based on the sensor output value of the rotation detection sensor that is not determined to have occurred.

Images