Motor control device, motor device

Abstract

A motor control device (10) is provided with, for example: a motor control block (11) that carries out feedback control of a driving current flowing in a motor (20); and a machine learning block (14) that detects a failure level of the motor (20) through analysis of input data including at least the driving current. The motor control block (11) preferably switches a control parameter or a control method dynamically according to the failure level. The input data may further include, for example, driving voltage applied to the motor (20). In addition, the input data may further include, for example, at least one of vibration and temperature of the motor (20) or a motor device (1) in which the motor is mounted.

Description

[Technical field]

[0001] The invention disclosed in this specification relates to a motor control device and a motor device using the same. [Background technology]

[0002] Motor control devices are installed in various motor devices (such as consumer equipment, industrial equipment, vehicles, and robots equipped with motors).

[0003] As an example of the related art, Patent Document 1 can be mentioned. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] JP 2020-022260 A Summary of the Invention [Problem to be solved by the invention]

[0005] However, conventional motor devices required regular inspection or replacement of the motor (including pre-replacement of non-faulty motors that do not necessarily require replacement) to prevent abnormal shutdowns of the motor control device due to motor failure, resulting in high maintenance costs (including the cost of inspecting or replacing the motor).

[0006] In view of the above-mentioned problems discovered by the inventors of the present application, the invention disclosed in this specification aims to provide a motor control device that can reduce the maintenance costs of a motor device, and a motor device using the same. [Means for solving the problem]

[0007] For example, the motor control device disclosed in this specification has a motor control block configured to perform feedback control of a drive current flowing through a motor, and a machine learning block configured to analyze input data including at least the drive current to detect a fault level of the motor, and the motor control block is configured to dynamically switch control parameters or control methods depending on the fault level. Effect of the Invention

[0008] According to the invention disclosed in this specification, it is possible to provide a motor control device that can reduce the maintenance costs of a motor device, and a motor device using the same. [Brief description of the drawings]

[0009] [Figure 1] FIG. 1 is a diagram illustrating a schematic configuration of a motor device. [Diagram 2] FIG. 2 is a diagram showing a first embodiment of a motor control block. [Diagram 3] FIG. 3 is a diagram showing a second embodiment of the motor control block. [Figure 4] FIG. 4 is a diagram illustrating a first operation example of the normalization block. [Diagram 5] FIG. 5 is a diagram illustrating a second operation example of the normalization block. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0010] <Motor device> 1 is a diagram showing a schematic configuration of a motor device. The motor device 1 of this configuration example includes a motor control device 10, a three-phase motor 20, a shunt resistor 30, a vibration sensor 40, and a temperature sensor 50.

[0011] The motor control device 10 is a semiconductor integrated circuit device (so-called motor control IC) that controls the rotational driving of the three-phase motor 20 by supplying three-phase driving currents Iu, Iv, and Iw, and three-phase driving voltages U, V, and W to the three-phase motor 20.

[0012] The three-phase motor 20 includes three-phase coils connected to the motor control device 10, and a rotor that rotates in response to drive currents Iu, Iv, and Iw flowing through the coils (all not shown). The rotational speed (angular velocity) of the rotor decreases as the frequencies of the drive currents Iu, Iv, and Iw decrease, and increases as the frequencies of the drive currents Iu, Iv, and Iw increase.

[0013] The shunt resistor 30 generates a current sense signal according to the current value of each of the drive currents Iu, Iv, and Iw. In the figure, for convenience of explanation, an example is shown in which the shunt resistor 30 is connected to the three-phase motor 20, but the shunt resistor 30 may be connected to a driver 12 described later. In addition, as a current detection method, a three-shunt method in which the drive currents Iu, Iv, and Iw are detected individually may be adopted, or a one-shunt method in which the drive currents Iu, Iv, and Iw are detected from the DC bus current of the driver 12 may be adopted.

[0014] The vibration sensor 40 is attached to, for example, the three-phase motor 20 (or any part of the motor device 1) and detects vibrations of the three-phase motor 20 (or the motor device 1). Note that, for example, an acceleration sensor or a gyro sensor can be suitably used as the vibration sensor 40.

[0015] The temperature sensor 50 is attached to, for example, the three-phase motor 20 (or any portion of the motor device 1) and detects the temperature of the three-phase motor 20 (or the motor device 1).

[0016] <Motor control device> The configuration and operation of the motor control device 10 will be described in detail with reference to Fig. 1. The motor control device 10 of this configuration example includes a motor control block 11, a driver 12, an ADC [analog-to-digital converter] 13, a machine learning block 14, and a normalization block 15.

[0017] The motor control block 11 receives input of digital current values ​​(corresponding to the current values ​​of the drive currents Iu, Iv, and Iw) from the ADC 13, and drives the driver 12 to perform feedback control of the drive currents Iu, Iv, and Iw flowing through the three-phase motor 20 so that the torque or rotational speed of the three-phase motor 20 matches a target value.

[0018] The motor control block 11 also has a function of dynamically switching the control parameters or control method according to the fault level detected by the machine learning block 14, a function of notifying the upper system of the fault level, and a function of communicating with the upper system. These functions will be described in detail later.

[0019] The driver 12 includes a three-phase half bridge (=three-phase upper FETs and lower FETs) connected to the motor control block 11, and generates three-phase drive currents Iu, Iv, and Iw flowing through the three-phase motor 20 based on three-phase gate signals (=upper gate signals applied to the gates of the three-phase upper FETs, and lower gate signals applied to the gates of the three-phase lower FETs) input from the motor control block 11. The driver 12 may be a separate IC externally attached to the motor control device 10.

[0020] The ADC 13 converts the analog current sense signal input from the shunt resistor 30 into a digital current value (corresponding to the current value of each of the drive currents Iu, Iv, and Iw) and outputs this to the motor control block 11 and the normalization block 15.

[0021] The machine learning block 14 analyzes the input data input from the normalization block 15 to detect the fault level (fault sign level) of the three-phase motor 20. The input data to the machine learning block 14 includes at least information on the drive currents Iu, Iv, and Iw, and in the example shown in the figure, further includes information on the vibration and temperature of the three-phase motor 20 (or the motor device 1). The input data to the machine learning block 14 may also include information on the drive voltages U, V, and W applied to the three-phase motor 20.

[0022] It is advisable to define multiple levels of failure, such as level 0 (safety) - level 4 (operating limit) and level 5 (failure).Failure levels are not necessarily proportional to the severity of the abnormality, and may be associated with the cause of the abnormality, such as level 1 being drive current abnormality and level 2 being drive voltage abnormality.

[0023] In addition, as the learning algorithm of the machine learning block 14, "supervised learning" that learns from labeled (answer-attached) data may be adopted, or "unsupervised learning" that learns from unlabeled data may be adopted. Of course, there is nothing to be avoided in adopting other existing algorithms such as "semi-supervised learning", "reinforcement learning", or "deep learning".

[0024] The normalization block 15 reduces the dependency on the rotation speed or control period of the three-phase motor 20 by normalizing the above-mentioned input data (drive current, drive voltage, vibration, temperature, etc.) before inputting it to the machine learning block 14. The purpose of introducing the normalization block 15 and its operation will be described in detail later.

[0025] <Dynamic control according to failure level> Next, the function of dynamically switching the control parameters or control method of the motor control block 11 according to the fault level detected by the machine learning block 14 will be described in detail.

[0026] For example, the motor control block 11 may have a table that associates fault levels with control parameters, and may select optimal control parameters for efficiently continuing to drive the three-phase motor 20 at the current fault level. Also, for example, the motor control block 11 may switch the control method from the vector control method to the 120-degree conduction method according to the fault level.

[0027] When the failure level reaches level 5 (failure), the motor control block 11 may forcibly stop driving the three-phase motor 20. If only one of the three phases is faulty, the current conduction mode may be switched from the three-phase current conduction mode to the two-phase current conduction mode to continue driving the three-phase motor 20.

[0028] In this way, the motor control block 11 performs dynamic control according to the fault level, so that the three-phase motor 20 that is starting to show signs of fault can be efficiently driven and the timing when replacement is really necessary (= the timing when the three-phase motor 20 reaches its operating limit) can be accurately notified to the upper system. As a result, it is possible to reduce the maintenance costs of the motor device 1 (= including the inspection costs or replacement costs of the three-phase motor 20).

[0029] The above-mentioned dynamic control may be performed autonomously by the motor control block 11 in response to the fault level detected by the machine learning block 14, or may be performed in response to a command from a higher-level system that reads the fault level via the motor control block 11.

[0030] <Notification to higher-level systems> Next, a notification function to a higher-level system in the motor control block 11 will be described. The motor control block 11 notifies an alert to the higher-level system, for example, via one interrupt line. The timing of the alert notification may be set, for example, when a change in the fault level occurs or when the fault level reaches a predetermined level.

[0031] The host system that receives the above-mentioned alert can read out the fault level stored in an internal register (not shown) of the motor control block 11 via the communication line. In addition, the host system can read out the fault level at any time even if it has not received an alert from the motor control block 11.

[0032] It is only necessary to store the latest fault level in the internal register, that is, the motor control block 11 only needs to update the value stored in the internal register when a change in the fault level occurs.

[0033] The internal register may also store supplemental information grasped by the motor control block 11, such as the cause of the change in the failure level and the degree of abnormality of each of the multiple input data (current, voltage, vibration, and temperature). In that case, not only information related to the drive current but also information related to the drive voltage, vibration, and temperature may be input to the motor control block 11.

[0034] <Motor control block (first embodiment)> 2 is a diagram showing a first embodiment of the motor control block 11. The motor control block 11 of this embodiment includes an excitation current control system 101, a calculator 102, a speed controller 103, a torque model 104, a three-phase / two-phase conversion unit 105, calculators 106 and 107, current controllers 108 and 109, an axis error detector 110, Lead angle control value The control circuit includes a setting unit 111, a calculator 112, a phase synchronization controller 113, a non-interference calculator (motor model) 114, calculators 115 and 116, a two-phase / three-phase conversion unit 117, and a rotor position detection unit 118, and performs feedback control of the drive currents Iu, Iv, and Iw by a three-phase sensorless full vector control method.

[0035] The excitation current control system 101 outputs a d-axis current command Id*.

[0036] The calculator 102 outputs the difference value between the angular velocity command ω1* and the angular velocity ω1.

[0037] The speed controller 103 (so-called ASR [automatic speed regulator]) outputs a torque command τ* for making the angular velocity ω1 follow the angular velocity command ω1* by PI [proportional-integral] control according to the output value (= the difference between the angular velocity command ω1* and the angular velocity ω1) of the calculator 102. The speed controller 103 may have a function of performing any error control when a state in which the output value of the calculator 102 exceeds a predetermined error determination threshold continues over multiple cycles.

[0038] The torque model 104 converts the torque command τ* into a q-axis current command Iq*.

[0039] The three-phase / two-phase conversion unit 105 converts the three-phase drive currents Iu, Iv, and Iw input from the current detection unit 16 into a two-phase d-axis current Id and a q-axis current Iq using a predetermined conversion algorithm (such as Clarke conversion and Park conversion). The current detection unit 16 corresponds to, for example, the ADC 13 and the shunt resistor 30 in FIG. 1.

[0040] The calculator 106 outputs the difference between the d-axis current command Id* and the d-axis current Id.

[0041] The calculator 107 outputs the difference between the q-axis current command Iq* and the q-axis current Iq.

[0042] A current controller (so-called ACR [automatic current regulator]) 108 outputs a d-axis voltage command Vd* for making the d-axis current Id follow the d-axis current command Id* by PI control according to the output value of the calculator 106 (= the difference between the d-axis current command Id* and the d-axis current Id).

[0043] The current controller 109 outputs a q-axis voltage command Vq* for making the q-axis current Iq follow the q-axis current command Iq* by PI control according to the output value of the calculator 107 (=the difference value between the q-axis current command Iq* and the q-axis current Iq).

[0044] The axis error detector 110 detects the axis error Δθ from the d-axis current Id, the q-axis current Iq, and the corrected d-axis voltage command Vd** and q-axis voltage command Vq**. With a configuration including such an axis error detector 110, an encoder for detecting the rotor position θ is not required, so that it is possible to reduce the cost of the motor device 1 and improve its reliability.

[0045] Lead angle control value setting unit 111 outputs a predetermined lead angle control setting value (for example, zero).

[0046] The calculator 112 outputs the difference between a predetermined lead angle control setting value and the axis error Δθ.

[0047] A phase lock controller 113 (so-called PLL [phase locked loop]) outputs an angular velocity ω1 (estimate) for converging the axis error Δθ to the lead angle control set value by PI control according to the output value (= the difference value between a predetermined lead angle control set value and the axis error Δθ) of the calculator 112. This makes it possible to absorb control delays, etc.

[0048] The non-interference calculator 114 generates a d-axis voltage correction value ΔVd* (=-ω1×Lq*×Iq*) and a q-axis voltage correction value ΔVq* (=ω1×Ld*×Id*+kE*×ω1) by non-interference calculation processing based on the d-axis current command Id*, the q-axis current command Iq*, and the angular velocity ω1. Note that Ld and Lq respectively indicate the d-axis coil inductance and the q-axis coil inductance, and kE indicates the back electromotive force constant.

[0049] The calculator 115 adds the d-axis voltage command Vd* and the d-axis voltage correction value ΔVd* together, and outputs the corrected d-axis voltage command Vd**.

[0050] The calculator 116 adds the q-axis voltage command Vq* and the q-axis voltage correction value ΔVq* together, and outputs the corrected q-axis voltage command Vq**.

[0051] A two-phase / three-phase conversion unit 117 converts the two-phase d-axis voltage command Vd** and the q-axis voltage command Vq** into three-phase (six in total, upper and lower) gate signals hu / lu , hv / lv and hw / lw and output to the driver 12.

[0052] Rotor position detection section 118 integrates angular velocity ω1 to detect rotor position θ, and outputs it to three-phase / two-phase conversion section 105 and two-phase / three-phase conversion section 117, respectively.

[0053] Here, the motor control block 11 dynamically switches at least one of the PI gain and error judgment threshold of the speed controller 103, the PI gains of each of the current controllers 108 and 109, the d-axis coil inductance Ld and q-axis coil inductance Lq and back electromotive force constant kE of the non-interference calculator 114, the coil resistance R and the d-axis coil inductance Ld and q-axis coil inductance Lq of the axis error detector 110, and the PI gain of the phase synchronization controller 113, depending on the fault level detected by the machine learning block 14 (not shown).

[0054] For example, if it is assumed in advance that the torque of the three-phase motor 20 will weaken as the fault level increases, it is advisable to perform auto-tuning to automatically change the PI gain of the speed controller 103 in proportion to the fault level.

[0055] Such dynamic control makes it possible to continue to drive the three-phase motor 20 efficiently even when the motor 20 begins to show signs of failure.

[0056] The control method of the motor control block 11 is not limited to the above-mentioned full vector control method, but may be another vector control method or may be a method other than the vector control method. In the following, a second embodiment other than the vector control method is proposed.

[0057] <Motor control block (second embodiment)> 3 is a diagram showing a second embodiment of the motor control block 11. The motor control block 11 of this embodiment includes a signal input unit 201, a speed detection unit 202, a calculator 203, a current command generation unit 204, a current selection unit 205, a calculator 206, a voltage command generation unit 207, a pulse width modulation unit 208, an open phase setting unit 209, and a conduction mode selection unit 210, and performs feedback control of the drive currents Iu, Iv, and Iw by a 120-degree conduction control method using three-phase Hall sensors.

[0058] The signal input unit 201 samples the three-phase Hall signals Hu, Hv, and Hw input from a Hall sensor (not shown) of the three-phase motor 20.

[0059] The speed detection unit 202 detects the angular speed ωr of the three-phase motor 20 based on the sampling output of the signal input unit 201 .

[0060] The calculator 203 outputs the difference between the angular velocity command ωr* and the angular velocity ωr.

[0061] The current command generating unit 204 (corresponding to a speed controller) outputs a current command I* for making the angular velocity ωr follow the angular velocity command ωr* by PI control according to the output value of the calculator 203 (=the difference value between the angular velocity command ωr* and the angular velocity ωr).

[0062] The current selection unit 205 outputs, as a selected drive current I, one of the drive currents Iu, Iv, and Iw flowing through the driver 12 in accordance with the current conduction mode selection signal VMODE.

[0063] The calculator 206 outputs the difference between the current command I* and the selected drive current I.

[0064] A voltage command generating unit 207 (corresponding to a current controller) outputs a voltage command Vo* by PI control according to the output value of the calculator 206 (=the difference between the current command I* and the selected drive current I).

[0065] The pulse width modulation unit 208 generates three-phase pulse width modulation signals PWMu, PWMv, and PWMw by performing distribution processing and pulse width modulation processing of the voltage command Vo* in a conduction mode (two-phase conduction mode or three-phase conduction mode) corresponding to the conduction mode selection signal VMODE.

[0066] The open phase setting unit 209 selects an open phase in response to the current-mode selection signal VMODE, and outputs three-phase (six in total, upper and lower) gate signals hu / lu, hv / lv, and hw / lw to the driver 12.

[0067] The conduction mode selection unit 210 generates a conduction mode selection signal VMODE based on the Hall signals Hu, Hv, and Hw.

[0068] Here, the motor control block 11 dynamically switches at least one of the PI gain of the current command generating unit 204 (corresponding to a speed controller) and the PI gain of the voltage command generating unit 207 (corresponding to a current controller) according to the fault level detected by the machine learning block 14 (not shown). Such dynamic control makes it possible to continue to efficiently drive the three-phase motor 20 that is beginning to show signs of fault.

[0069] <canonicalization block> The following describes in detail the significance of introducing the normalization block 15 and its operation (normalization method). If the normalization block 15 is not introduced, even a change in the rotation speed or control period (control frequency) of the three-phase motor 20 would cause a change in behavior in the input data (drive current, drive voltage, vibration, temperature, etc.) to the machine learning block 14, which could result in erroneous detection of the fault level.

[0070] Therefore, the motor control device 10 includes a normalization block 15 that normalizes the above input data before inputting it to the machine learning block 14 to reduce the dependency on the rotation speed or control period of the three-phase motor 20.

[0071] With this configuration, even if the rotation speed or control period of the three-phase motor 20 changes, the fault level can be accurately detected without changing the weights of the machine learning block 14 in response to those changes.

[0072] FIG. 4 is a diagram showing a first operation example of the normalization block 15, and depicts, from top to bottom, a sensor output signal SO(@ωx) at a rotational speed ωx (e.g., equivalent to the output signal of a vibration sensor), a normalized sensor output signal SO(n), and a sensor output signal SO(@ωy) at a rotational speed ωy (<ωx).

[0073] As shown in the figure, the pulse periods Tx and Ty of the sensor output signal SO vary depending on the rotation speeds ωx and ωy of the three-phase motor 20. Therefore, by variably controlling the sampling period of the sensor output signal SO in accordance with the rotation speeds ωx and ωy of the three-phase motor 20, the pulse periods Tx and Ty of the sensor output signal SO are normalized to a predetermined fixed length Tn, making it possible to accurately detect the fault level by the machine learning block 14 having uniform weights, regardless of the rotation speeds ωx and ωy of the three-phase motor 20.

[0074] 5 is a diagram showing a second operation example of the normalization block 15, and shows a power spectrum PS obtained by performing frequency conversion processing such as FFT [fast Fourier transform] and DCT [discrete cosine transform] on the above-mentioned sensor output signal SO. Specifically, from the right, the power spectrum PS(@ωx) of the sensor output signal SO at a rotation speed ωx, the normalized power spectrum PS(n), and the power spectrum PS(@ωy) of the sensor output signal SO at a rotation speed ωy are depicted.

[0075] As shown in the figure, the peak frequencies fx and fy of the power spectrum PS vary depending on the rotation speeds ωx and ωy of the three-phase motor 20. Therefore, if the peak frequencies fx and fy of the power spectrum PS are normalized to a predetermined fixed frequency fn according to the rotation speeds ωx and ωy of the three-phase motor 20, it becomes possible to accurately detect the fault level by the machine learning block 14 having uniform weights, regardless of the rotation speeds ωx and ωy of the three-phase motor 20.

[0076] <Summary> The various embodiments described above will now be summarized.

[0077] For example, the motor control device disclosed in this specification has a motor control block configured to perform feedback control of a drive current flowing through a motor, and a machine learning block configured to analyze input data including at least the drive current to detect a fault level of the motor, and the motor control block is configured to dynamically switch control parameters or control methods depending on the fault level (first configuration).

[0078] In the motor control device having the above first configuration, the input data may further include a drive voltage to be applied to the motor (second configuration).

[0079] In addition, in the motor control device having the first or second configuration described above, the input data may be configured (third configuration) to further include at least one of vibration and temperature of the motor or a motor device equipped with the motor.

[0080] In the motor control device having any one of the first to third configurations, the motor control block may be configured to perform feedback control of the drive current by a vector control method (fourth configuration).

[0081] Furthermore, in the motor control device having the above-mentioned fourth configuration, the motor control block may be configured (fifth configuration) to dynamically switch at least one of the PI gain and error judgment threshold of the speed controller, the PI gain of the current controller, the d-axis coil inductance and q-axis coil inductance and back electromotive force constant of the non-interference calculator, the coil resistance and d-axis coil inductance and q-axis coil inductance of the axis error detector, and the PI gain of the phase synchronization controller in response to the fault level.

[0082] In the motor control device having any one of the first to third configurations, the motor control block may be configured to perform feedback control of the drive current by a 120-degree conduction control method (sixth configuration).

[0083] In addition, in the motor control device having the sixth configuration described above, the motor control block may be configured (seventh configuration) to dynamically switch at least one of the PI gain of the speed controller and the PI gain of the current controller in response to the fault level.

[0084] In addition, a motor control device having any of the above-mentioned first to seventh configurations may be configured (eighth configuration) to further have a normalization block configured to normalize the input data before inputting it to the machine learning block, thereby reducing dependency on the rotational speed or control period of the motor.

[0085] In the motor control device having any one of the first to eighth configurations above, the motor control block may have a function of notifying a host system of the fault level (ninth configuration).

[0086] Furthermore, the motor device disclosed in this specification has a configuration (tenth configuration) including a motor and a motor control device having any one of the first to ninth configurations described above.

[0087] <Other Modifications> In addition to the above-mentioned embodiment, various technical features disclosed in this specification can be modified in various ways without departing from the spirit of the technical creation. In other words, the above-mentioned embodiment should be considered to be illustrative and not restrictive in all respects, and the technical scope of the present invention should not be limited to the above-mentioned embodiment, but should be understood to include all modifications that fall within the meaning and scope equivalent to the scope of the claims. [Industrial Applicability]

[0088] The motor control device disclosed in this specification can be used as a means for controlling a variety of motor devices. [Explanation of symbols]

[0089] 1 Motor Device 10 Motor control device 11 Motor Control Block 12 Drivers 13 ADC 14 Machine Learning Blocks 15 Normalization Blocks 16 Current detection section 20 Three-phase motor 30 Shunt Resistor 40 Vibration Sensor 50 Temperature Sensor 101 Excitation current control system 102 Arithmetic unit 103 Speed ​​Controller 104 Torque Model 105 Three-phase / two-phase conversion unit 106, 107 Arithmetic unit 108, 109 Current controller 110 Axis Error Detector 111 Lead angle control value setting unit 112 Arithmetic unit 113 Phase Synchronization Controller 114 Non-interference calculator (motor model) 115, 116 Arithmetic unit 117 Two-phase / three-phase converter 118 Rotor position detector 201 Signal input section 202 Speed ​​detection unit 203 Arithmetic unit 204 Current command generation section 205 Current selection section 206 Arithmetic unit 207 Voltage command generator 208 Pulse Width Modulation Section 209 Open phase setting section 210 Power supply mode selection section

Claims

1. A motor control block configured to perform feedback control of a drive current flowing through the motor; a machine learning block configured to analyze input data including at least the drive current to detect a fault level of the motor; having the motor control block dynamically switches a control method from a vector control method to a 120-degree conduction method, dynamically switches a conduction mode from a three-phase conduction mode to a two-phase conduction mode, or dynamically switches at least one of a PI gain of a speed controller and a PI gain of a current controller, depending on the fault level; Motor control device.

2. A motor control block configured to perform feedback control of a drive current flowing through a motor using a vector control method; a machine learning block configured to analyze input data including at least the drive current to detect a fault level of the motor; having the motor control block dynamically switches at least one of a PI gain and an error determination threshold of a speed controller, a PI gain of a current controller, a d-axis coil inductance and a q-axis coil inductance and a back electromotive force constant of a non-interference calculator, a coil resistance and a d-axis coil inductance and a q-axis coil inductance of an axis error detector, and a PI gain of a phase synchronization controller according to the fault level; Motor control device.

3. A motor control block configured to perform feedback control of a drive current flowing through a motor using a 120-degree conduction control method; a machine learning block configured to analyze input data including at least the drive current to detect a fault level of the motor; having the motor control block dynamically switches at least one of a PI gain of a speed controller and a PI gain of a current controller in response to the fault level; Motor control device.

4. The input data further includes a drive voltage to be applied to the motor. The motor control device according to any one of claims 1 to 3.

5. The input data further includes at least one of vibration and temperature of the motor or a motor device having the motor mounted thereon. The motor control device according to any one of claims 1 to 4.

6. A normalization block configured to normalize the input data before inputting the input data to the machine learning block to reduce dependency on a rotation speed or a control period of the motor. The motor control device according to any one of claims 1 to 5.

7. The normalization block normalizes a pulse period of the input data to a predetermined fixed length, or normalizes a peak frequency of a power spectrum obtained by performing a frequency conversion process on the input data to a predetermined fixed frequency. The motor control device according to claim 6.

8. The input data includes vibration of the motor or a motor device equipped with the motor. The motor control device according to claim 7.

9. The motor control block has a function of notifying a host system of the fault level. The motor control device according to any one of claims 1 to 8.

10. A motor; A motor control device according to any one of claims 1 to 9, A motor device having a

Images