notebook.community

Edit and run

Time series to irradiance tutorial





In [1]:



    


%matplotlib inline
import sys
import os
#sys.path.append('..') #append the parent directory to the system path
import pvlib
import pandas as pd 
import matplotlib.pyplot as plt
try:
    import seaborn
    seaborn.set_context('paper')
except ImportError:
    pass



    











In [2]:



    


datapath = os.path.join(pvlib.__path__[0], 'test', 'data')
fname = '703165TY.csv' #Use absolute path if the file is not in the local directory
TMY, meta = pvlib.tmy.readtmy3(filename=os.path.join(datapath, fname))
TMY.index.name = 'Time'
meta['SurfTilt'] = 30
meta['SurfAz'] = 0
meta['Albedo'] = 0.2

# create pvlib Location object
sand_point = pvlib.location.Location(meta['latitude'], meta['longitude'], tz='US/Alaska', 
                                     altitude=meta['altitude'], name=meta['Name'].replace('"',''))
print(sand_point)

# TMY data seems to be given as hourly data with time stamp at the end
# shift the index 30 Minutes back for calculation of sun postions
TMY = TMY.shift(freq='-30Min')



    


















    






SAND POINT: latitude=55.317, longitude=-160.517, tz=US/Alaska, altitude=7.0

Retreive required parameters

Calculate the solar position for all times in the TMY file. Requires either PyEphem or a compiled spa_c library.





In [3]:



    


solpos = pvlib.solarposition.get_solarposition(TMY.index, sand_point, method='pyephem')
solpos.head()



    


















    
Out[3]:










  
    

      
      apparent_elevation
      apparent_azimuth
      elevation
      azimuth
      apparent_zenith
      zenith
    

    

      Time
      
      
      
      
      
      
    

  
  
    

      1997-01-01 00:30:00-09:00
      -54.750103
       328.861096
      -54.750103
       328.861096
       144.750103
       144.750103
    

    

      1997-01-01 01:30:00-09:00
      -57.530205
       353.265280
      -57.530205
       353.265280
       147.530205
       147.530205
    

    

      1997-01-01 02:30:00-09:00
      -56.627712
        18.751764
      -56.627712
        18.751764
       146.627712
       146.627712
    

    

      1997-01-01 03:30:00-09:00
      -52.329549
        41.456666
      -52.329549
        41.456666
       142.329549
       142.329549
    

    

      1997-01-01 04:30:00-09:00
      -45.717933
        60.026503
      -45.717933
        60.026503
       135.717933
       135.717933
    

  




















In [4]:



    


# append solar position data to TMY DataFrame
TMY['SunAz'] = solpos['apparent_elevation']
TMY['SunZen'] =  solpos['apparent_zenith']

TMY['HExtra'] = pvlib.irradiance.extraradiation(doy=TMY.index.dayofyear)

TMY['AM'] = pvlib.atmosphere.relativeairmass(z=TMY.SunZen)

DFOut = pvlib.clearsky.disc(Time=TMY.index,GHI=TMY.GHI, SunZen=TMY.SunZen)

TMY['DNI_gen_DISC'] = DFOut['DNI_gen_DISC']
TMY['Kt_gen_DISC'] = DFOut['Kt_gen_DISC']
TMY['AM'] = DFOut['AM']
TMY['Ztemp'] = DFOut['Ztemp']



    


















    






---------------------------------------------------------------------------
ValueError                                Traceback (most recent call last)
<ipython-input-4-0523b848c9cd> in <module>()
      5 TMY['HExtra'] = pvlib.irradiance.extraradiation(doy=TMY.index.dayofyear)
      6 
----> 7 TMY['AM'] = pvlib.atmosphere.relativeairmass(z=TMY.SunZen)
      8 
      9 DFOut = pvlib.clearsky.disc(Time=TMY.index,GHI=TMY.GHI, SunZen=TMY.SunZen)

/home/will/git_repos/pythonic-PVLIB/pvlib/atmosphere.pyc in relativeairmass(z, model)
    227   Expect={'z': ('array','num','x<=90','x>=0'),
    228       'model': ('default','default=kastenyoung1989')}
--> 229   var=pvt.Parse(Vars,Expect)
    230 
    231   if ('kastenyoung1989') == var.model.lower():

/home/will/git_repos/pythonic-PVLIB/pvlib/pvl_tools.pyc in __init__(self, dct, Expect)
     78     '''
     79     def __init__(self, dct, Expect):
---> 80         self.__dict__.update(self.parse_fcn(dct,Expect))
     81 
     82     def parse_fcn(self, kwargs, Expect):

/home/will/git_repos/pythonic-PVLIB/pvlib/pvl_tools.pyc in parse_fcn(self, kwargs, Expect)
    196                                     raise Exception('Error: Numeric input "'+arg+' " fails on logical test " '+ re.findall(reg,string)[0]+'"')
    197                             except:
--> 198                                 if not(eval(lambdastring)(kwargs[arg])):
    199                                     raise Exception('Error: Numeric input "'+arg+' " fails on logical test " '+ re.findall(reg,string)[0]+'"')
    200 

ValueError: The truth value of an array with more than one element is ambiguous. Use a.any() or a.all()

Perez





In [4]:



    


models = ['Perez', 'Hay-Davies', 'Isotropic', 'King', 'Klucher', 'Reindl']
TMY['Perez'] = pvlib.pvl_perez(SurfTilt=meta['SurfTilt'],
                               SurfAz=meta['SurfAz'],
                               DHI=TMY.DHI,
                               DNI=TMY.DNI,
                               HExtra=TMY.HExtra,
                               SunZen=TMY.SunZen,
                               SunAz=TMY.SunAz,
                               AM=TMY.AM)

HayDavies





In [5]:



    


TMY['Hay-Davies'] = pvlib.pvl_haydavies1980(SurfTilt=meta['SurfTilt'],
                                            SurfAz=meta['SurfAz'],
                                            DHI=TMY.DHI,
                                            DNI=TMY.DNI,
                                            HExtra=TMY.HExtra,
                                            SunZen=TMY.SunZen,
                                            SunAz=TMY.SunAz)

Isotropic





In [6]:



    


TMY['Isotropic'] = pvlib.pvl_isotropicsky(SurfTilt=meta['SurfTilt'],
                                          DHI=TMY.DHI)

King Diffuse model





In [7]:



    


TMY['King'] = pvlib.pvl_kingdiffuse(SurfTilt=meta['SurfTilt'],
                                    DHI=TMY.DHI,
                                    GHI=TMY.GHI,
                                    SunZen=TMY.SunZen)

Klucher Model





In [8]:



    


TMY['Klucher'] = pvlib.pvl_klucher1979(SurfTilt=meta['SurfTilt'],
                                       SurfAz=meta['SurfAz'],
                                       DHI=TMY.DHI,
                                       GHI=TMY.GHI,
                                       SunZen=TMY.SunZen,
                                       SunAz=TMY.SunAz)

Reindl





In [9]:



    


TMY['Reindl'] = pvlib.pvl_reindl1990(SurfTilt=meta['SurfTilt'],
                                     SurfAz=meta['SurfAz'],
                                     DHI=TMY.DHI,
                                     DNI=TMY.DNI,
                                     GHI=TMY.GHI,
                                     HExtra=TMY.HExtra,
                                     SunZen=TMY.SunZen,
                                     SunAz=TMY.SunAz)



    











In [10]:



    


yearly = TMY[models].resample('A', how='sum').squeeze() / 1000.0  # kWh
monthly = TMY[models].resample('M', how='sum', kind='period') / 1000.0
daily = TMY[models].resample('D', how='sum') / 1000.0

Plot Results





In [12]:



    


ax = TMY[models].plot(title='Modelled in-plane diffuse irradiance')
# ax.legend(ncol=3, loc='upper center', title='In-plane diffuse irradiance')
ax.set_ylim(0, 800)
ylabel = ax.set_ylabel('Diffuse Irradiance [W]')



    


















    
























In [13]:



    


TMY[models].describe()



    


















    
Out[13]:










  
    

      
      Perez
      Hay-Davies
      Isotropic
      King
      Klucher
      Reindl
    

  
  
    

      count
       4531.000000
       8760.000000
       8760.000000
       8760.000000
       8760.000000
       8760.000000
    

    

      mean
        126.020357
         58.782821
         56.830698
         64.360708
         64.622931
         59.137119
    

    

      std
         98.944032
         87.587632
         85.451560
         91.309309
         95.026610
         88.152787
    

    

      min
          0.715740
          0.000000
          0.000000
          0.000000
          0.000000
          0.000000
    

    

      25%
         58.800900
          0.000000
          0.000000
          0.000000
          0.000000
          0.000000
    

    

      50%
         96.567972
          5.616401
          5.598076
          6.155411
          5.667071
          5.622082
    

    

      75%
        173.452541
         89.248955
         84.904156
        103.102595
         98.899346
         89.601183
    

    

      max
        558.131136
        531.675901
        526.219164
        533.076491
        533.061174
        535.251475
    

  




















In [14]:



    


TMY[models].dropna().plot(kind='density')



    


















    
Out[14]:








<matplotlib.axes._subplots.AxesSubplot at 0x7f50b79a52d0>










    
























In [15]:



    


ax_daily = daily[models].plot(title='Daily diffuse irradiation')
ylabel = ax_daily.set_ylabel('Irradiation [kWh]')



    


















    
























In [16]:



    


ax_monthly = monthly[models].plot(title='Monthly diffuse irradiation', kind='bar')
ylabel = ax_monthly.set_ylabel('Irradiation [kWh]')



    


















    
























In [17]:



    


yearly.plot(kind='barh')



    


















    
Out[17]:








<matplotlib.axes._subplots.AxesSubplot at 0x7f50b7712e10>

Compute the mean deviation from average for each model and display as a function of the model





In [18]:



    


mean_yearly = yearly.mean()
yearly_mean_deviation = (yearly - mean_yearly) / yearly * 100.0
yearly_mean_deviation.plot(kind='bar')



    


















    
Out[18]:








<matplotlib.axes._subplots.AxesSubplot at 0x7f50b5a5dad0>










    
























In [ ]:

Content source: rubennj/pvlib-python

Similar notebooks:

notebook.community | gallery | about