solarspatialtools.cmv.compute_cmv

solarspatialtools.cmv.compute_cmv(timeseries, positions, reference_id=None, method='jamaly', options=None)

Find Cloud Motion Vector based on clear sky index timeseries from a cluster of sensors.

Default uses the method by Jamaly and Kleissl [1]. An alternate method described by Gagne [2] is available. Optionally computes relative to a single reference point rather than a global computation across all possible site pairs.

Parameters

timeseriespandas.DataFrame

dataframe of kt timeseries for all sensors. Columns should be labelled with the sensor id. All sensors should share a uniform time index.

positionspandas.DataFrame

dataframe of the site positions. Index must be the site IDs. The first column should be the x coordinate of each site, and the second column should be the y coordinate. Coordinates must represent a rectilinear coordinate system, measured in meters. Consider converting to UTM using spatial.latlon2utm().

reference_idnumeric, str, iterable, or None (default None)

The identifier of a single reference site id within the dataframe. OR a list/tuple of identifiers of references within the dataframe. OR None to signify that all possible site pairs should be considered.

methodstr (default ‘jamaly’)

Method to use. Currently accepted methods are ‘jamaly’ and ‘gagne’

optionsdict (default {})

Dictionary of detailed QC arguments for the methods.

All Methods:

ktlimfloat (default 0.4)

Minimum permissible clear sky index for time period

Jamaly:

minvelocityfloat (default 0 m/s)

Minimum permissible pairwise velocity in m/s

maxvelocityfloat (default 70 m/s)

Maximum permissible pairwise velocity in m/s

var_s_limfloat (default 0.05)

Minimum permissible pairwise variation ratio in signals

mincorrfloat (default 0.8)

Minimum permissible correlation coefficient for site pairs

Gagne:

No method specific options

Returns

cmv_velnumeric

The cloud motion vector magnitude in meters per second

cmv_thetanumeric

The cloud motion direction as an angle measured CCW from east in rads

outdataWindspeedData

An object containing detailed data about the individual sites

[1] M. Jamaly and J. Kleissl, “Robust cloud motion estimation by spatio-temporal correlation analysis of irradiance data,” Solar Energy, vol. 159, pp. 306–317, Jan. 2018. [Online]. Available: https://www.sciencedirect.com/science/article/pii/S0038092X17309556

[2] A. Gagné, N. Ninad, J. Adeyemo, D. Turcotte, and S. Wong, “Directional Solar Variability Analysis,” in 2018 IEEE Electrical Power and Energy Conference (EPEC) (2018) pp. 1–6, iSSN: 2381-2842 https://www.researchgate.net/publication/330877949_Directional_Solar_Variability_Analysis