| CPC G06F 16/2365 (2019.01) [G05B 23/0229 (2013.01)] | 18 Claims |

|
1. A data processing apparatus, comprising:
at least one processor;
a data acquisition unit that uses the at least one processor to acquire time-series data related to running of an apparatus;
a stop acquisition unit that uses the at least one processor to acquire a stop of the running of the apparatus;
a calculation unit that uses the at least one processor to calculate an amount of variation by calculating a difference in a value of the data in a period before a stop period and a period after the stop period of the apparatus based on the data acquired by the data acquisition unit and the stop of the running of the apparatus acquired by the stop acquisition unit; and
a correction unit that uses the at least one processor to correct, after changing a reference value according to the amount of variation in the data between before and after the stop period of the apparatus, the data using the reference value; and
the correction unit supplies the calculated difference to the learning processing unit as corrected process data;
a learning processing unit that uses the at least one processor to generate a machine learning model using the data corrected by the correction unit for detecting whether there is an anomality in the process data;
wherein
the stop period is a maintenance period for the apparatus, and
operations of the apparatus are controlled using control data generated by the machine learning model.
|