pecos.pv module¶

The pv module contains custom methods for PV applications.

pecos.pv.insolation(G, tfilter=None, per_day=True)[source]¶

Compute insolation defined as:

\(H=\int{Gdt}\)

where \(G\) is irradiance and \(dt\) is the time step between observations. The time integral is computed using the trapezoidal rule. Results are given in [irradiance units]*seconds.

Parameters:

G : pd.DataFrame

Irradiance time series

tfilter : pd.Series (optional)

Time filter containing boolean values for each time index

per_day : boolean (optional)

Flag indicating if the results should be computed per day, default = True

Returns:

H : pd.DataFrame

Insolation

pecos.pv.energy(P, tfilter=None, per_day=True)[source]¶

Convert energy defined as:

\(E=\int{Pdt}\)

where \(P\) is power and \(dt\) is the time step between observations. The time integral is computed using the trapezoidal rule. Results are given in [power units]*seconds.

Parameters:

P : pd.DataFrame

Power time series

tfilter : pd.Series (optional)

Time filter containing boolean values for each time index

per_day : boolean (optional)

Flag indicating if the results should be computed per day, default = True

Returns:

E : pd.DataFrame

Energy

pecos.pv.performance_ratio(E, H_poa, P_ref, G_ref=1000)[source]¶

Compute performance ratio defined as:

\(PR=\dfrac{Y_{f}}{Yr} = \dfrac{\dfrac{E}{P_{ref}}}{\dfrac{H_{poa}}{G_{ref}}}\)

where \(Y_f\) is the observed energy (AC or DC) produced by the PV system (kWh) divided by the DC power rating at STC conditions. \(Y_r\) is the plane-of-array insolation (kWh/m2) divided by the reference irradiance (1000 W/m2).

Parameters:

E : pd.DataFrame with a single column or pd.Series

Energy (AC or DC)

H_poa : pd.DataFrame with a single column or pd.Series

Plane of array insolation

P_ref : float

DC power rating at STC conditions

G_ref : float (optional)

Reference irradiance, default = 1000

Returns:

PR : pd.DataFrame

Performance ratio

pecos.pv.normalized_current(I, G_poa, I_sco, G_ref=1000)[source]¶

Compute normalized current defined as:

\(NI = \dfrac{\dfrac{I}{I_{sco}}}{\dfrac{G_{poa}}{G_{ref}}}\)

where \(I\) is current, \(I_{sco}\) is the short circuit current at STC conditions, \(G_{poa}\) is the plane-of-array irradiance, and \(G_{ref}\) is the reference irradiance.

Parameters:

I : pd.DataFrame with a single column or pd.Series

Current

G_poa : pd.DataFrame with a single column or pd.Series

Plane of array irradiance

I_sco : float

Short circuit current at STC conditions

G_ref : float (optional)

Reference irradiance, default = 1000

Returns:

NI : pd.DataFrame

Normalized current

pecos.pv.normalized_efficiency(P, G_poa, P_ref, G_ref=1000)[source]¶

Compute normalized efficiency defined as:

\(NE = \dfrac{\dfrac{P}{P_{ref}}}{\dfrac{G_{poa}}{G_{ref}}}\)

where \(P\) is the observed power (AC or DC), \(P_{ref}\) is the DC power rating at STC conditions, \(G_{poa}\) is the plane-of-array irradiance, and \(G_{ref}\) is the reference irradiance.

Parameters:

P : pd.DataFrame with a single column or pd.Series

Power (AC or DC)

G_poa : pd.DataFrame with a single column or pd.Series

Plane of array irradiance

P_ref : float

DC power rating at STC conditions

G_ref : float (optional)

Reference irradiance, default = 1000

Returns:

NE : pd.DataFrame

Normalized efficiency

pecos.pv.performance_index(E, E_predicted)[source]¶

Compute performance index defined as:

\(PI=\dfrac{E}{\hat{E}}\)

where \(E\) is the observed energy from a PV system and \(\hat{E}\) is the predicted energy over the same time frame. \(\hat{E}\) can be computed using by first predicting power using pecos.pv.basic_pvlib_performance_model or methods in pvlib.pvsystem and then convert power to energy using pecos.pv.enery.

Unlike with the performance ratio, the performance index should be very close to 1 for a well functioning PV system and should not vary by season due to temperature variations.

Parameters:

E : pd.DataFrame with a single column or pd.Series

Observed energy

E_predicted : pd.DataFrame with a single column or pd.Series

Predicted energy

Returns:

PI : pd.DataFrame

Performance index

pecos.pv.energy_yield(E, P_ref)[source]¶

Compute energy yield is defined as:

\(EY=\dfrac{E}{P_{ref}}\)

where \(E\) is the observed energy from a PV system and \(P_{ref}\) is the DC power rating of the system at STC conditions.

Parameters:

E : pd.DataFrame with a single column or pd.Series

Observed energy

P_ref : float

DC power rating at STC conditions

Returns:

EY : pd.DataFrame

Energy yield

pecos.pv.clearness_index(H_dn, H_ea)[source]¶

Compute clearness index defined as:

\(Kt=\dfrac{H_{dn}}{H_{ea}}\)

where \(H_{dn}\) is the direct-normal insolation (kWh/m2) \(H_{ea}\) is the extraterrestrial insolation (kWh/m2) over the same time frame. Extraterrestrial irradiation can be computed using pvlib.irradiance.extraradiation. Irradiation can be converted to insolation using pecos.pv.insolation.

Parameters:

H_dn : pd.DataFrame with a single column or pd.Series

Direct normal insolation

H_ea : pd.DataFrame with a single column or pd.Series

Extraterrestrial insolation

Returns:

Kt : pd.DataFrame

Clearness index

pecos.pv.basic_pvlib_performance_model(parameters, latitude, longitude, wind_speed, air_temp, poa_global, poa_diffuse=None, model='SAPM')[source]¶

Compute a very basic pv performance model using the SAPM or single diode model from pvlib. Input includes observed wind speed, air temperature, and POA irradiance. Default model options, defined in pvlib, are used to compute the performance model. Use pvlib directly to customize the model.

Parameters:

parameters : dict

Model parameters, see pvlib.pvsystem module for more details

latitude : float

Latitude

longitude : float

Longitude

wind speed : pd.DataFrame with a single column or pd.Series

Wind speed time series

air_temp : pd.DataFrame with a single column or pd.Series

Air temperature time series

poa_global : pd.DataFrame with a single column or pd.Series

Global POA irradiance time series

poa_diffuse : pd.DataFrame with a single column or pd.Series (optional)

Diffuse POA irradiance time series, default = 0

model : string (optional)

‘SAPM’ or ‘singlediode’, default = ‘SAPM’

Returns:

model : pd.DataFrame

Predicted Isc, Imp, Voc, Vmp