adap

PyPI

This is python package used detect different types of anomalies in compressors' sensors time series data.

Versions collected1

Direct dependencies0From release 0.0.1

Last observed20 Oct, 19:03 UTC

README

Version 0.0.1 documentation

This is a simple example package. You can use GitHub-flavored Markdown to write your content.

Anomaly type	Type code	Description	Example in the literature and/or alternative terminology	Importance ranking (with respect to time series visualization in this case study)	Potential end-user and application
Impossible values	A	Values 6666 are highly impossible in compressor sensor data and consider as out of range value of the sensors.	-	Important, but it can be deleted easily using simple rule	-
Constant values within the acceptable region for long time period	B	Constant values within the acceptable region for long time period (more than one day)	-	Important can be detected using rate of change	-
Large sudden spike	C	The spikes are larger than the surrounding data points but less than 6666. Example one or two values much higher than previous or next values	-	At any point in the time series, used 3 times standard deviation as threshold	-
Drift	D	Gradual change in values in positive to zero	-	High priority (Emergency shutdown, ESD)	-
Sudden zero	E	Value suddenly shifts in values in positive to zero	-	High priority (Emergency shutdown, ESD)	-

Table 1: Load current related anomaly types

Anomaly type	Type code	Description	Example in the literature and/or alternative terminology	Importance ranking (with respect to time series visualization in this case study)	Potential end-user and application
Impossible values	A	Values 6666 are highly impossible in compressor sensor data and consider as out of range value of the sensors.	-	Important, but it can be deleted easily using simple rule	-
Constant values within the acceptable region for long time period	B	Constant values within the acceptable region for long time period (more than one day)	-	Important can be detected using rate of change	-
Large sudden spike	C	The spikes are larger than the surrounding data points but less than 6666. Example one or two values much higher than previous or next values	-	At any point in the time series, used 3 times standard deviation as threshold	-
Drift	D	Gradual change in values in positive to zero	-	High priority (Emergency shutdown, ESD)	-
Sudden zero	E	Value suddenly shifts in values in positive to zero	-	High priority (Emergency shutdown, ESD)	-
Step up (sudden shift)	F	Increase the mean while keeping the variance constant	-	High priority (Emergency shutdown, ESD)	-
Step down (sudden shift)	G	Decrease the mean while keeping the variance constant	-	Normal, indicate motor flips from OFF to ON	-

Table 2: Suction pressure related anomaly types

Anomaly type	Type code	Description	Example in the literature and/or alternative terminology	Importance ranking (with respect to time series visualization in this case study)	Potential end-user and application
Impossible values	A	Values 6666 are highly impossible in compressor sensor data and consider as out of range value of the sensors.	-	Important, but it can be deleted easily using simple rule	-
Constant values within the acceptable region for long time period	B	Constant values within the acceptable region for long time period (more than one day)	-	Important can be detected using rate of change	-
Large sudden spike	C	The spikes are larger than the surrounding data points but less than 6666. Example one or two values much higher than previous or next values	-	At any point in the time series, used 3 times standard deviation as threshold	-
Drift	D	Gradual change in values in positive to zero	-	High priority (Emergency shutdown, ESD)	-
Sudden zero	E	Value suddenly shifts in values in positive to zero	-	High priority (Emergency shutdown, ESD)	-
Step up (sudden shift)	F	Increase the mean while keeping the variance constant	-	Normal, indicate motor flips from OFF to ON	-
Step down (sudden shift)	G	Decrease the mean while keeping the variance constant	-	High priority (Emergency shutdown, ESD)	-

Table 2: Discharge pressure related anomaly types

Awaiting observations

Download counts and popularity will appear here as the collector receives them.

Collected releases

Version	Published	Direct dependencies	Open version
0.0.1	23 Mar 2024	0

No matching advisories collected

This describes our collected data; it is not a security assessment.

PENDING

Shared repository data · GitHub

Meaningful changes

Current collected state Current

Observed since 20 Oct, 19:03 UTC

No superseded metadata states collected. Unchanged observations do not create new revisions.

package:6102cf3f609e48aaf813d315ac471461c08df9ab82317bca37141274764f1289