Machine learning device which learns current command for motor, motor controller, and machine learning method

1. A machine learning device configured to learn a current command for a motor, the machine learning device comprising:

a state observation unit configured to observe state variables including

(a) a motor rotation speed or

(b) a motor torque command of the motor, and

at least one of

(c) a motor torque,

(d) a motor current, and

(e) a motor voltage of the motor; and

a learning unit configured to learn the current command for the motor based on the state variables observed by the state observation unit, wherein

the learning unit includes a reward calculation unit configured to calculate a reward based on an output of the state observation unit,

the learning unit is configured to update, based on the calculated reward, a value function relating to the current command for the motor, and

the reward is increased or decreased in response to at least one of the motor torque, the motor current, or the motor voltage observed by the state observation unit.

2. The machine learning device according to claim 1 , further comprising a decision unit configured to determine the current command for the motor based on an output of the learning unit.

3. The machine learning device according to claim 1 , wherein

the motor is a three-phase alternating-current synchronous motor, and

the current command for the three-phase alternating-current synchronous motor includes a d-axis current command and a q-axis current command which are electrically and magnetically orthogonal to each other.

4. The machine learning device according to claim 3 , wherein learning of the d-axis current command and the q-axis current command is performed in terms of a rotor phase angle of the three-phase alternating-current synchronous motor.

5. The machine learning device according to claim 1 , wherein

for each motor rotation speed or motor torque command for the motor, the reward calculation unit is configured to

increase the reward when the motor torque increases,

decrease the reward when the motor current exceeds a predetermined value, and

decrease the reward when the motor voltage is saturated.

6. A machine learning device configured to learn a current command for a motor, the machine learning device comprising:

a state observation unit configured to observe state variables including

a motor rotation speed or a motor torque command of the motor, and

at least one of a motor torque, a motor current, and a motor voltage of the motor; and

a learning unit configured to learn the current command for the motor based on the state variables observed by the state observation unit,

wherein the learning unit includes:

an error calculation unit configured to calculate an error based on the output of the state observation unit and supervised data; and

an error model update unit configured to update an error model relating to the current command for the motor based on the output of the state observation unit and an output of the error calculation unit.

7. The machine learning device according to claim 1 , wherein the machine learning device further comprises a neural network.

8. The machine learning device according to claim 1 , wherein the machine learning device is configured to mutually share or exchange the current command for the motor as learned by the machine learning device among a plurality of machine learning devices via a communication medium.

9. The machine learning device according to claim 1 , wherein the machine learning device is located on at least one server in a cloud.

10. A motor controller, comprising:

the machine learning device according to claim 1 ;

a motor drive unit configured to drive and control the motor; and

a motor control unit configured to control the current command for the motor through the motor drive unit based on the current command for the motor as learned by the machine learning device.

11. The motor controller according to claim 10 , wherein

the state observation unit is configured to observe both the motor current and the motor voltage, and

the motor torque is calculated from the motor current and the motor voltage as observed.

12. A machine learning method of learning a current command for a motor, the machine learning method comprising:

observing state variables including

(a) a motor rotation speed or

(b) a motor torque command of the motor, and

at least one of

(c) a motor torque,

(d) a motor current, and

(e) a motor voltage for the motor; and

learning the current command for the motor based on the state variables as observed, wherein

the learning of the current command for the motor includes:

calculating a reward based on the state variables as observed; and

updating a value function relating to the current command for the motor based on the state variables as observed and the reward as calculated.

13. The machine learning method according to claim 12 , wherein in the reward calculation, for each motor rotation speed or motor torque command,

the reward is increased when the motor torque increases,

the reward is decreased when the motor current exceeds a predetermined value, and

the reward is decreased when the motor voltage is saturated.

14. A machine learning method of learning a current command for a motor, the machine learning method comprising:

observing state variables including

a motor rotation speed or a motor torque command of the motor, and

at least one of a motor torque, a motor current, and a motor voltage for the motor; and

learning the current command for the motor based on the state variables as observed, wherein

the learning of the current command for the motor includes:

calculating an error based on the state variables as observed and supervised data; and

updating an error model relating to the current command for the motor based on the state variables as observed and the error as calculated.