ENH: Implement PVSyst Cell Temperature Model

[mikofski]

PVSyst Cell Temperature Model
For users using PVSyst model chain, there is no equivalent cell temperature model. The only cell temperature model is the Sandia array performance model, pvlib.pvsystem.sapm_celltemp. The PVSyst cell temperature model is different.

Describe the solution you'd like
Add a corresponding method like pvlib.pvsystem.pvsyst_celltemp. I'll bet it's already implemented in PVLIB_MATLAB, right?

Describe alternatives you've considered
leave it to the user to implement

def pvsyst_celltemp(ambient_temp, gti, panel_efficiency, windspeed, alpha_absorption=0.9,
                    natural_convenction_coeff=29, forced_convection_coeff=0):
    """
    Calculate PVSyst cell temperature.

    Parameters
    ------------
    ambient_temp : numeric
        Ambient temperature in degrees Celsius
    gti : numeric
        Global tilted irradiance, AKA incident irradiance in the plane of array (POA) in W/m^2
    panel_efficiency : numeric
        PV module efficiency as a fraction
    windspeed : numeric
        Wind speed in m/s
    alpha_absorption : numeric
        Absorption coefficient, default is 0.9
    natural_convenction_coeff : numeric
        Natural convection coefficient, default is 29
    forced_convection_coeff : numeric
        Forced convection coefficient, default is zero

    Returns
    --------
    celltemp : numeric
        Cell temperature in degrees Celsius
    """
    combined_convection_coeff = (
        natural_convenction_coeff + forced_convection_coeff * windspeed
    )
    return (ambient_temp
        + alpha_absorption * gti * (1 - panel_efficiency) / combined_convection_coeff)

I'm not married to these names, they are only for demonstration

Additional context

• is this related to Add PVsyst as option to ModelChain #487 at all? I tried to look for any corresponding or existing issues
• if this cell temperature model is implmented then should others be as well? what does SAM use?

[cwhanse]

Great proposal @mikofski. The PVSyst temperature model is an extension of the Faiman temperature model. We don't have either in PVLib MATLAB. They are easy enough to code but I don't know of a database of coefficients. Like the Sandia model, coefficient values depend both on the module construction and its mounting. PVSyst provides some defaults.

I read the recent bulletin board exchange between Mike Anderson and Andre Mermoud, it appears that there is still uncertainty regarding some of the PVsyst model terms (efficiency).

The Faiman model is prescribed by at least one IEC standard so it is of broad interest. I would like to see this model added as well. Different pull request, though.

How do people feel about changing the naming convention: celltemp_sapm, celltemp_pvsyst etc. rather than sapm_celltemp etc.?

[wholmgren]

At the risk of issue scope creep... what about a celltemp.py module with functions sapm, pvsyst, etc.? Maybe better to address this in #436.

[cwhanse]

Works for me

…

Sent from my iPhone On Aug 31, 2018, at 4:27 PM, Will Holmgren <notifications@github.com<mailto:notifications@github.com>> wrote: At the risk of issue scope creep... what about a celltemp.py module with functions sapm, pvsyst, etc.? — You are receiving this because you commented. Reply to this email directly, view it on GitHub<#552 (comment)>, or mute the thread<https://github.com/notifications/unsubscribe-auth/AFJNLzMx4IT5Yf_PAQkfKxA9NJR_1sRFks5uWbhggaJpZM4WVsLl>.

[wholmgren]

Closed by #628.

[adriesse]

I have a PR coming which I would like to attach to this issue.