Time-series data processing device, time-series data processing system, and time-series data processing method
AssigneeMitsubishi Electric Corporation
InventorTakaaki Nakamura
Life of this patent
Estimating the term…
Abstract
An event waveform extracting unit (3) extracts an event waveform from time-series data. A co-occurrence rate calculating unit (4) calculates co-occurrence rates of event waveforms among the time-series data. A grouping unit (5) classifies the time-series data into groups depending the co-occurrence rates of the event waveforms. An event information generating unit (6) determines the time at which the periods during which event waveforms occur overlap with each other among the time-series data included in each group, and generates event information identifying an event related to the event waveforms on the basis of the determined time.
Claims14
A time-series data processing device comprising:
- processing circuitry to:
- extract waveform data estimated to be changed because of an event having occurred in subject equipment from each of a plurality of time-series data sequentially observed over time from the subject equipment;
- calculate a plurality of co-occurrence rates of the waveform data extracted among time-series data;
- classify the time-series data into groups depending on the plurality of co-occurrence rates of the waveform data calculated and an associated matrix, the associated matrix being formed on a basis of results of comparing the plurality of co-occurrence rates of the waveform data to a threshold;
- and
- determine time at which periods of occurrence of the waveform data overlap with each other among the time-series data included in a group into which the time-series data are classified, and generate event information identifying an event related to the waveform data on a basis of the determined time.
A time-series data processing system comprising:
- processing circuitry to:
- extract waveform data estimated to be changed because of an event having occurred in subject equipment from each of a plurality of time-series data sequentially observed over time from the subject equipment;
- calculate a plurality of co-occurrence rates of the waveform data extracted among the time-series data;
- classify the time-series data into groups depending on the plurality of co-occurrence rates of the waveform data calculated and an associated matrix, the associated matrix being formed on a basis of results of comparing the plurality of co-occurrence rates of the waveform data to a threshold;
- determine time at which periods of occurrence of the waveform data overlap with each other among the time-series data included in a group into which the time-series data are classified, and generate event information identifying an event related to the waveform data on a basis of the determined time;
- detect outlier data from the time-series data, the outlier data being partial string data falling within an abnormality range;
- and
- determine abnormality of the subject equipment on a basis of the outlier data detected and the event information generated.
A time-series data processing method comprising:
- extracting waveform data estimated to be changed because of an event having occurred in subject equipment from each of a plurality of time-series data sequentially observed over time from the subject equipment;
- calculating a plurality of co-occurrence rates of the waveform data extracted among the time-series data;
- classifying the time-series data into groups depending on the plurality of co-occurrence rates of the waveform data calculated and an associated matrix, the associated matrix being formed on a basis of results of comparing the plurality of co-occurrence rates of the waveform data to a threshold;
- and
- determining time at which periods of occurrence of the waveform data overlap with each other among the time-series data included in a group into which the time-series data are classified, and generating event information identifying an event related to the waveform data on a basis of the determined time.
Dependencies are inferred from the claim text.