Temperature models#

pvlib provides a variety of models for predicting the operating temperature of a PV module from irradiance and weather inputs. These models range from simple empirical equations requiring just a few multiplications to more complex thermal balance models with numerical integration.

Types of models#

Temperature models predict one of two quantities:

module temperature: the temperature as measured at the back surface of a PV module. Easy to measure, but usually marginally less than the cell temperature which determines efficiency.
cell temperature: the temperature of the PV cell itself. The relevant temperature for PV modeling, but almost never measured directly.

Temperature models estimate these quantities using inputs like incident irradiance, ambient temperature, and wind speed. Each model also takes a set of parameter values that represent how a PV module responds to those inputs.

Parameter values generally depend on both the PV module technologies, the mounting configuration of the module, and on any weather parameters that are not included in the model. Note that, despite models conventionally being associated with either cell or module temperature, it is actually the parameter values that determine which of the two temperatures are predicted, as they will produce the same type of temperature from which they were originally derived.

Another aspect of temperature models is whether they account for the thermal inertia of a PV module. Temperature models are either:

steady-state: the module is assumed to have been at the specified operating conditions for a sufficiently long time for its temperature to reach equilibrium.
transient: the module’s thermal inertia is included in the model, causing a lag in modeled temperature change following changes in the inputs.

Other effects that temperature models may consider include the photoconversion efficiency and radiative cooling.

The temperature models currently available in pvlib are summarized in the following table:

Model	Type	Transient?	Weather inputs
Model	Type	Transient?	POA irradiance	Ambient temperature	Wind speed	Downwelling IR [1]
`faiman()`	either		✓	✓	✓
`faiman_rad()`	either		✓	✓	✓	✓
`fuentes()`	either	✓	✓	✓	✓
`generic_linear()`	either		✓	✓	✓
`noct_sam()`	cell		✓	✓	✓
`pvsyst_cell()`	cell		✓	✓	✓
`ross()`	cell		✓	✓
`sapm_cell()`	cell		✓	✓	✓
`sapm_module()`	module		✓	✓	✓

In addition to the core models above, pvlib provides several other functions for temperature modeling:

prilliman(): an “add-on” model that reprocesses the output of a steady-state model to apply transient effects.
sapm_cell_from_module(): a model for estimating cell temperature from module temperature.

Model parameters#

Some temperature model functions provide default values for their parameters, and several additional sets of temperature model parameter values are available in pvlib.temperature.TEMPERATURE_MODEL_PARAMETERS. However, these generic values may not be suitable for all modules and mounting configurations. It should be noted that using the default parameter values for each model generally leads to different modules temperature predictions. This alone does not mean one model is better than another; it’s just evidence that the measurements used to derive the default parameter values were taken on different PV systems in different locations under different conditions.

Parameter values for one model (e.g. u0, u1 for faiman()) can be converted to another model (e.g. u_c, u_v for pvsyst_cell()) using GenericLinearModel.

Module-specific values can be obtained via testing, for example following the IEC 61853-2 standard for the Faiman model; however, such values still do not capture the dependency of temperature on system design and other variables.

Currently, pvlib provides no functionality for fitting parameter values using measured temperature.