Working with Time Series¶

To demonstrate working with time series, we notice that SensorTimeSeries is an extension of CogniteTimeSeries in the WindTurbine model

In [8]:

from wind_turbine import WindTurbineClient

from wind_turbine import WindTurbineClient

In [9]:

pygen = WindTurbineClient.from_toml("config.toml")

pygen = WindTurbineClient.from_toml("config.toml")

Listing timeseries¶

We can list the timeseries directly with the sensor_time_series attribute

In [10]:

timeseries = pygen.sensor_time_series.list(limit=5)
timeseries

timeseries = pygen.sensor_time_series.list(limit=5) timeseries

Out[10]:

	space	external_id	aliases	concept_id	description	is_step	name	source_unit	standard_name	type_	data_record
0	sp_wind	V52-WindTurbine.Gear_D_Z	[Gearbox Displacement]	http://purl.org/aspect/gearbox_displacement	Gearbox Displacement Z (laser sensor) (not cal...	False	Gear_D_Z	V	gearbox_displacement	numeric	{'version': 1, 'last_updated_time': 2024-11-16...
1	sp_wind	V52-WindTurbine.MyA4_O	[Edgewise Bending Moment]	http://purl.org/aspect/edgewise_bending_moment	Edgewise Bend. Mom. Blade A R6.6m Offset	False	MyA4_O	kNm	edgewise_bending_moment	numeric	{'version': 1, 'last_updated_time': 2024-11-16...
2	sp_wind	V52-WindTurbine.cnt0	[Low Speed Shaft Rotational Speed]	http://purl.org/aspect/low_speed_shaft_rotatio...	Revolution per minutes of the low speed shaft,...	False	cnt0	RPM	low_speed_shaft_rotational_speed	numeric	{'version': 1, 'last_updated_time': 2024-11-16...
3	sp_wind	witznitz_efficiency	[Witznitz Efficiency]	http://purl.org/aspect/solarPanel/efficiency	Witznitz Efficiency	False	Witznitz Efficiency	percent	Witznitz Efficiency	numeric	{'version': 1, 'last_updated_time': 2024-11-16...
4	sp_wind	V52-WindTurbine.MxA4_O	[Flapwise Bending Moment]	http://purl.org/aspect/flapwise_bending_moment	Flapwise Bend. Mom. Blade A R6.6m Offset	False	MxA4_O	kNm	flapwise_bending_moment	numeric	{'version': 1, 'last_updated_time': 2024-11-16...

Retriving Data Points¶

To retrieve data points we use the .select() method in the same way as when we do regular queries.

For example, if we want to view the rotor_speed_controller of the rotor for the hornsea_1_mill_1 turbine:

In [11]:

df = (
    pygen.wind_turbine.select()
    .name.equals("hornsea_1_mill_1")
    .rotor.rotor_speed_controller.data.retrieve_dataframe(limit=100)
)
df

df = ( pygen.wind_turbine.select() .name.equals("hornsea_1_mill_1") .rotor.rotor_speed_controller.data.retrieve_dataframe(limit=100) ) df

Out[11]:

	NodeId(sp_wind, V52-WindTurbine.ROT)
2018-01-01 16:00:00.020	23.6999
2018-01-01 16:00:00.040	23.6999
2018-01-01 16:00:00.060	23.6999
2018-01-01 16:00:00.080	23.6999
2018-01-01 16:00:00.100	23.6999
...	...
2018-01-01 16:00:01.920	24.0995
2018-01-01 16:00:01.940	24.0995
2018-01-01 16:00:01.960	24.0995
2018-01-01 16:00:01.980	24.0995
2018-01-01 16:00:02.000	24.0995

100 rows × 1 columns

Retrieving pandas DataFrame in Time Zone¶

In [14]:

from zoneinfo import ZoneInfo

from zoneinfo import ZoneInfo

In [15]:

tz = ZoneInfo("Europe/Oslo")

tz = ZoneInfo("Europe/Oslo")

In [18]:

df = (
    pygen.wind_turbine.select()
    .name.equals("hornsea_1_mill_1")
    .rotor.rotor_speed_controller.data.retrieve_dataframe(limit=1_000, timezone=tz)
)
df

df = ( pygen.wind_turbine.select() .name.equals("hornsea_1_mill_1") .rotor.rotor_speed_controller.data.retrieve_dataframe(limit=1_000, timezone=tz) ) df

Out[18]:

	NodeId(sp_wind, V52-WindTurbine.ROT)
2018-01-01 17:00:00.020000+01:00	23.6999
2018-01-01 17:00:00.040000+01:00	23.6999
2018-01-01 17:00:00.060000+01:00	23.6999
2018-01-01 17:00:00.080000+01:00	23.6999
2018-01-01 17:00:00.100000+01:00	23.6999
...	...
2018-01-01 17:00:19.920000+01:00	23.9998
2018-01-01 17:00:19.940000+01:00	23.9998
2018-01-01 17:00:19.960000+01:00	23.9998
2018-01-01 17:00:19.980000+01:00	23.9998
2018-01-01 17:00:20+01:00	23.9998

1000 rows × 1 columns

Plotting Timeseries¶

pandas has built in plotting and thus just chose a retrive methods that returns a pandas DataFrame and call .plot() on it.

In [19]:

df.plot(figsize=(15, 7));

df.plot(figsize=(15, 7));

In [ ]: