Device and method for correcting thermal displacement using generated data collection plan

1 . A control device for controlling a machine by generating a data collection plan formed of a combination of operating conditions for acquiring data from the machine in operation, the control device comprising:

a processor configured to

set a specification of the machine,

extract a constrain condition associated with an operation of the machine, the constrain condition including a number of operations,

generate a plurality of operating conditions including a set of values of parameters associated with the set specification based on the specification and the constrain condition,

generate a data collection plan from the generated plurality of operating conditions, wherein the processor is configured to

in response to a number of the plurality of operating conditions included in the data collection plan to be generated exceeding a limit of the number of operations included in the extracted constrain condition, further thin out the plurality of operating conditions, and

in response to the number of the plurality of operating conditions included in the data collection plan to be generated being fewer than the limit of the number of operations, further iteratively generate the plurality of operating conditions within the limit of the number of operations; and

correct thermal displacement of the machine, by using machine learning predicting a value for thermal displacement correction based on the generated data collection plan.

2 . The control device according to claim 1 , wherein

the processor is further configured to

select the parameters to be included in the plurality of operating conditions of the machine based on the specification and the constrain condition, and

generate the plurality of operating conditions including the set of values of the parameters selected based on the specification and the constrain condition.

3 . A data collection plan generating method conducted by a control device for controlling a machine by generating a data collection plan formed of a combination of operating conditions for acquiring data from the machine in operation, the data collection plan generating method comprising:

a specification setting procedure for setting a specification of the machine;

a constrain condition extracting procedure for extracting a constrain condition associated with an operation of the machine, the constrain condition including a number of operations;

an operation condition generating procedure for generating a plurality of operating conditions including a set of values of parameters associated with the specification set in the specification setting procedure based on the specification and the constrain condition;

a data collection plan generating procedure for generating a data collection plan from the plurality of operating conditions generated in the operation condition generating procedure, wherein the data collection plan generating method includes

in response to a number of the plurality of operating conditions included in the data collection plan to be generated exceeding a limit of the number of operations included in the extracted constrain condition, further thinning out the plurality of operating conditions, and

in response to the number of the plurality of operating conditions included in the data collection plan to be generated being fewer than the limit of the number of operations, further iteratively generating the plurality of operating conditions within the limit of the number of operations; and

a procedure for correcting thermal displacement of the machine by using machine learning predicting a value for thermal displacement correction based on the generated data collection plan.

4 . The data collection plan generating method according to claim 3 , wherein

after the constrain condition extracting procedure, a parameter selecting procedure is further conducted to select the parameters to be included in the plurality of operating conditions of the machine based on the specification and the constrain conditions, and

the operating condition generating procedure is conducted to generate the plurality of operating conditions including the set of values of the parameters selected in the parameter selecting procedure based on the specification and the constrain conditions.