US 11,853,052 B2
Cluster based classification for time series data
Markus M. Geipel, Munich (DE); Nikou Günnemann-Gholizadeh, Munich (DE); Stephan Merk, Munich (DE); and Sebastian Mittelstädt, Augsburg (DE)
Assigned to SIEMENS AKTIENGESELLSCHAFT, Munich (DE)
Appl. No. 17/436,141
Filed by Siemens Aktiengesellschaft, Munich (DE)
PCT Filed Feb. 17, 2020, PCT No. PCT/EP2020/054130
§ 371(c)(1), (2) Date Sep. 3, 2021,
PCT Pub. No. WO2020/178016, PCT Pub. Date Sep. 10, 2020.
Claims priority of application No. 19160577 (EP), filed on Mar. 4, 2019.
Prior Publication US 2022/0128987 A1, Apr. 28, 2022
Int. Cl. G05B 23/02 (2006.01)
CPC G05B 23/0281 (2013.01) [G05B 23/0221 (2013.01); G05B 23/0272 (2013.01)] 20 Claims
OG exemplary drawing
 
1. A device for analyzing time series data, comprising:
a processor; and
memory;
wherein the device is configured to:
obtain time series data monitored on a machine;
segment the time series data into multiple time segments;
determine a cluster of time segments estimated to have time characteristics which match time characteristics of the monitored time series data;
provide cluster information to a classification stage as a first input;
check whether label information indicating classification information associated with at least one time segment of the time segments is available;
provide the label information to the classification stage as a second input to classify the cluster based on the label information associated with at least one of the time segments if label information indicating the classification information associated with at least one time segment of the time segments is available;
present at least a part of the time series data of the cluster to a user if none of the time segments of the cluster has associated label information; and
classify the cluster and generate label information associated with the time segments of the cluster based on a user input received in response to said presentation of at least a part of the time series data, the generated label information indicating a result of classifying the cluster.