Control apparatus, control method, and storage medium

Patent metadata

Field	Value
Publication number	US-11960267-B2
Application number	US-202117223044-A
Country	US
Kind code	B2
Filing date	Apr 6, 2021
Priority date	Apr 24, 2020
Publication date	Apr 16, 2024
Grant date	Apr 16, 2024

Abstract

Provided is a control apparatus including an acquisition unit configured to acquire a measurement value measured regarding control target equipment, a first control unit configured to output an operation amount of the control target equipment according to the measurement value by at least one of feedback control or feed-forward control, a second control unit configured to output an operation amount of the control target equipment according to the measurement value using a model learnt by using learning data, and a switching unit configured to perform switching between the first control unit and the second control unit by which the control target equipment is controlled.

First claim

What is claimed is:

1. A control apparatus comprising:
an acquisition unit configured to acquire a measurement value measured regarding control target equipment;
a first control unit configured to output an operation amount of the control target equipment according to the measurement value by at least one of feedback control or feed-forward control;
a second control unit configured to output an operation amount of the control target equipment according to the measurement value using a model learnt by using learning data; and
a switching unit configured to perform switching between the first control unit and the second control unit by which the control target equipment is controlled,
wherein the switching unit is configured to perform the switching according to a difference between the measurement value and a target value, and
wherein the switching unit is configured to perform the switching from the control by the first control unit to the control by the second control unit in response to a state occurring multiple times where the difference between the measurement value and the target value is higher than a reference value within a reference time window.

2. The control apparatus according to claim 1 , wherein:
the first control unit is configured to output an operation amount calculated based on the measurement value and a target value; and
the model of the second control unit is learnt by using learning data that includes measurement data including the measurement value and the operation amount of the control target equipment, and is configured to output a recommended operation amount of the control target equipment for increasing a reward value determined by a previously set reward function in response to input of the measurement data.

3. The control apparatus according to claim 2 , wherein:
the reward function is a function by which the reward value is increased as the measurement value gets closer to the target value; and
the switching unit is configured to perform the switching on the basis of a comparison result between a threshold based on the target value and the measurement value.

4. The control apparatus according to claim 3 , wherein the switching unit is configured to
set a threshold used for switching from the control by the second control unit to the control by the first control unit to be higher than a threshold used for switching from the control by the first control unit to the control by the first second control unit.

5. The control apparatus according to claim 1 , wherein the switching unit is configured to cause
the second control unit to perform the control until reference time elapses from start of the control on the control target equipment, and
the first control unit to perform the control after the reference time has elapsed from the start of the control on the control target equipment.

6. The control apparatus according to claim 1 , wherein the first control unit is configured to perform feedback control using at least one of proportional control, integral control, or differential control.

7. The control apparatus according to claim 1 , wherein:
the first control unit is operable in
an auto mode for calculating and outputting, in response to input of the measurement value, an operation amount of the control target equipment according to the measurement value, and
a manual mode for outputting, in response to input of the operation amount that is to be output, the operation amount;
the second control unit is configured to input the operation amount of the control target equipment to the first control unit; and
the switching unit is configured to perform the switching by switching a mode of the first control unit.

8. The control apparatus according to claim 7 , wherein in a case where the switching from the manual mode to the auto mode is performed, the first control unit is configured to control the operation amount before and after the switching such that an abrupt change of the operation amount before and after the switching is suppressed.

9. A control method comprising:
acquiring a measurement value measured regarding control target equipment;
outputting an operation amount of the control target equipment according to the measurement value by at least one of feedback control or feed-forward control in a first control phase;
outputting an operation amount of the control target equipment according to the measurement value using a model learnt by using learning data in a second control phase; and
performing switching between the first control phase and the second control phase by which the control target equipment is controlled,
wherein the performing switching is performed according to a difference between the measurement value and a target value, and
wherein the performing switching is performed by switching from control in accordance to the first control phase to control in accordance to the second control phase in response to a state occurring multiple times where the difference between the measurement value and the target value is higher than a reference value within a reference time window.

10. A storage medium storing a control program for causing a computer to function as:
an acquisition unit configured to acquire a measurement value measured regarding control target equipment;
a first control unit configured to output an operation amount of the control target equipment according to the measurement value by at least one of feedback control or feed-forward control;
a second control unit configured to output an operation amount of the control target equipment according to the measurement value using a model learnt by using learning data; and
a switching unit configured to perform switching between the first control unit and the second control unit by which the control target equipment is controlled,
wherein the switching unit is configured to perform the switching according to a difference between the measurement value and a target value, and
wherein the switching unit is configured to perform the switching from the control by the first control unit to the control by the second control unit in response to a state occurring multiple times where the difference between the measurement value and the target value is higher than a reference value within a reference time window.

11. The control apparatus according to claim 1 , wherein the switching unit performs control switching from the control by the second control unit to the control by the first control unit in response to a state where the difference is lower than the reference value.

Assignees

• Yokogawa Electric Corp

Inventors

• Takami Go
• Nishijo Fumio
• Tanaka Satoru

Classifications

• G05B19/4142 Primary

  characterised by the use of a microprocessor (G05B19/4141 takes precedence) · CPC title

• B25J9/161

  Hardware, e.g. neural networks, fuzzy logic, interfaces, processor · CPC title

• B25J9/163

  learning, adaptive, model based, rule based expert control · CPC title

• G05B13/0265 Primary

  the criterion being a learning criterion · CPC title

• G06N20/00

  Machine learning · CPC title